Rounding up the usual suspects a standard target-gene approachfor resolving the interfamilial phylogenetic relationships of
ecribellate orb-weaving spiders with a new family-rank classification(Araneae Araneoidea)
Dimitar Dimitrova Ligia R Benavidesbc Miquel A Arnedocd Gonzalo GiribetcCharles E Griswolde Nikolaj Scharfff and Gustavo Hormigab
aNatural History Museum University of Oslo PO Box 1172 Blindern NO-0318 Oslo Norway bDepartment of Biological Sciences The George
Washington University Washington DC 20052 USA cMuseum of Comparative Zoology amp Department of Organismic and Evolutionary Biology
Harvard University 26 Oxford Street Cambridge MA 02138 USA dDepartament de Biologia Animal and Institut de Recerca de la Biodiversitat
(IRBio) Universitat de Barcelona Avinguda Diagonal 643 Barcelona 08071 Catalonia Spain eArachnology California Academy of Sciences 55
Music Concourse Drive Golden Gate Park San Francisco CA 94118 USA fCenter for Macroecology Evolution and Climate Natural History
Museum of Denmark University of Copenhagen Universitetsparken 15 Copenhagen DK-2100 Denmark
Accepted 19 March 2016
Abstract
We test the limits of the spider superfamily Araneoidea and reconstruct its interfamilial relationships using standard molecularmarkers The taxon sample (363 terminals) comprises for the first time representatives of all araneoid families including the firstmolecular data of the family Synaphridae We use the resulting phylogenetic framework to study web evolution in araneoids Ara-neoidea is monophyletic and sister to Nicodamoidea rank n Orbiculariae are not monophyletic and also include the RTA cladeOecobiidae and Hersiliidae Deinopoidea is paraphyletic with respect to a lineage that includes the RTA clade Hersiliidae andOecobiidae The cribellate orb-weaving family Uloboridae is monophyletic and is sister group to a lineage that includes the RTAClade Hersiliidae and Oecobiidae The monophyly of most Araneoidea families is well supported with a few exceptions Anapidaeincludes holarchaeids but the family remains diphyletic even if Holarchaea is considered an anapid The orb-web is ancient havingevolved by the early Jurassic a single origin of the orb with multiple ldquolossesrdquo is implied by our analyses By the late Jurassic theorb-web had already been transformed into different architectures but the ancestors of the RTA clade probably built orb-webs Wealso find further support for a single origin of the cribellum and multiple independent losses The following taxonomic and nomen-clatural changes are proposed the cribellate and ecribellate nicodamids are grouped in the superfamily Nicodamoidea rank n
(Megadictynidae rank res and Nicodamidae stat n) Araneoidea includes 17 families with the following changes Araneidae is re-circumscribed to include nephilines Nephilinae rank res Arkyidae rank n Physoglenidae rank n Synotaxidae is limited to thegenus Synotaxus Pararchaeidae is a junior synonym of Malkaridae (syn n) Holarchaeidae of Anapidae (syn n) and Sinopimoidaeof Linyphiidae (syn n)copy The Willi Hennig Society 2016
Introduction
The orb-weaving spiders (ldquoOrbiculariaerdquo) includeat least one of the most diverse branches of thespider tree of lifemdashAraneoidea More than 12 500
species (approximately 28 of the more than 45 000described spider species) have been classified as mem-bers of one of the former 21 extant ldquoorbicularianrdquofamilies Although the defining trait of orbiculariansas their name suggests is the orb-web itself webarchitecture in this putative lineage is extraordinarilyvariable (Fig 1) ranging from the well-known bidi-mensional highly geometric snare with a framed set
Corresponding authorsE-mail addresses dimitardgwugmailcom and hormigagwuedu
CladisticsCladistics 33 (2017) 221ndash250
101111cla12165
copy The Willi Hennig Society 2016
of radii and a sticky spiral (eg in TetragnathidaeFig 6F) to highly irregular tridimensional webs (egin Linyphiidae Fig 6D G H) Almost everythingin between these architectural extremes seems to existand most of this web diversity is still undiscoveredor undocumented (eg Scharff and Hormiga 2012)In some cases foraging webs have been abandonedaltogether such as in the pirate spiders (MimetidaeFig 4C)
Two groups of orb-weaversmdashdeinopoids and arane-oidsmdashbuild similar webs that differ significantly in thestructure and composition of the silk of their capturespiral Traditionally regarded as a lineage these twogroups are now hypothesized not to form a clade(Dimitrov et al 2013 Bond et al 2014 Fernandezet al 2014) Deinopoids (Deinopidae Uloboridae) usecribellate silk for their sticky spiral (Fig 1A B) whilethe allegedly homologous counterpart in araneoids is
(A) (B)
(C) (D)
Fig 1 (A) The cribellate web of Sybota sp (Uloboridae) from Chile (DSC_2250) (B) The cribellate ogre-face spider Deinopis sp (Deinopidae)from Australia (DSC_0983) (C) The ecribellate Nephila plumipes building its orb-web Australia the highly reflective silk lines in this image arethe viscid capture spiral turns covered with a sticky glycoprotein a synapomorphy of Araneoidea The less reflective silk lines in between stickyturns are part of the temporary nonsticky spiral which in Nephila and its relatives are left in the finished web (DSC_6451) (D) Progradungulaotwayensis (Gradungulidae) from Australia with its ladder cribellate web an example of an early-branching araneomorph that illustrates theantiquity of cribellate silk (DSC_1424) Photos G Hormiga
222 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
made of a type of viscid silk that is unique to arane-oids (eg Fig 1C) Cribellate silk is ancient (egFig 1D)mdashit evolved in the early araneomorph lin-eagesmdashand thus sharing such type of silk among dei-nopoid taxa is expected to be symplesiomorphic Thistype of silk is spun by a spinning plate (the cribellum)in combination with a combing structure on the fourthleg metatarsus consisting of a row of modifiedmacrosetae (the calamistrum) Cribellate silk is ldquodryrdquoand is formed of thousands of fine looped fibrilswoven on a core of two axial fibres (eg Opell 1998fig 1) Its adhesive properties are attained by van derWaals and hygroscopic forces (Hawthorn and Opell2003) In contrast araneoids produce a novel type ofsticky silk in which the axial fibres are coated with aviscid glycoprotein This type of composite stickythread is produced faster presumably more economi-cally and attains a much higher stickiness than thedry deinopoid cribellate silk A large body of empiricalwork has studied and compared the biological andphysicochemical properties of these types of silks (seereview in Blackledge 2012)There is a marked disparity in species richness
between cribellate and ecribellate orb-weavers Themajority of orb-weaving spiders are members of thesuperfamily Araneoidea (the ecribellate orb-weavers 17families more than 12 000 species described) In com-parison Deinopoidea the cribellate orb-weaversinclude only 331 described species in two families Nico-damidae a small Austral group (29 species named) withboth cribellate and ecribellate members appears to bephylogenetically related to the ecribellate orb-weavers(Blackledge et al 2009 Dimitrov et al 2012) Thisasymmetry in species diversity between deinopoids andaraneoids has been attributed to the shift in type of cap-ture thread from dry fuzzy cribellate silk (Fig 1B) toviscid sticky silk (Fig 1C) combined with changes inthe silk spectral reflective properties and a transitionfrom horizontal to vertical orb-webs (references summa-rized in Hormiga and Griswold 2014) However recentstudies (Dimitrov et al 2013 Bond et al 2014Fernandez et al 2014) and the results presented hereshow that the contrast DeinopoideandashAraneoidea is nolonger valid and it is likely that evolution of webs anddiversification into new ecological niches are responsiblefor the differences in diversity of these spider clades (egDimitrov et al 2012)The question of whether cribellate and ecribellate
orb-webs can be traced to a single origin or haveevolved independently began to be debated in the 19thCentury (summarized in Coddington 1986) and hasbeen discussed extensively in the literature It was notuntil the late 1980s that a consensus began to emergeon the answer to this problem During the last threedecades the combination of comparative behaviouraldata (such as the seminal work of Eberhard 1982) and
cladistic approaches to analyse the available evidencehas favoured a monophyletic origin of orb-webs andthe monophyly of Orbiculariae (eg Levi and Cod-dington 1983 Coddington 1986 1990) with the pre-ponderance of evidence supporting this view comingfrom the webs and the concomitant stereotypical beha-viours used to build them Most research in the lasttwo decades has supported a single origin of the orb-web Because the monophyly of orb-weavers has beensupported primarily by behavioural and spinningorgan characters it has been challenging to test thepossibility that orb-webs were not convergent in thecribellate and ecribellate orb-weavers without referringto the building behaviours and silk products Geneticdata have played an increasingly important role inresolving spider phylogenetic relationships mostly inthe form of nucleotide sequences from a few genes (thenuclear and mitochondrial rRNA genes 18S 28S 12Sand 16S and a handful of protein-encoding genes fromwhich the most commonly used are the nuclear histoneH3 and the mitochondrial COI) often humorouslydescribed as ldquothe usual suspectsrdquo However the suc-cess of these markers as an independent test to resolveorbicularian relationships has been limited particularlyat the interfamilial level (eg Blackledge et al 2009Dimitrov et al 2012)Only one phylogenetic analysis of molecular data
with a sufficiently dense taxon sample to properlyaddress interfamilial relationships has recovered Orbic-ulariae as a clade albeit without support (Dimitrovet al 2012) Furthermore these nucleotide data failedto resolve or provide support for the relationshipsamong most orbicularian families the majority of deepinternodes are short Although most phylogeneticanalyses of DNA sequence data have found that orbic-ularians are not monophyletic this particular resulthas often been dismissed as ldquoartefactualrdquo (eg due totaxon sampling effects) or ldquomisleadingrdquomdashsuch hasbeen the convincing power of the orbicularian mono-phyly hypothesis For example in an analysis of thespider sequences available in GenBank Agnarssonet al (2013) explicitly stated that the placement ofUloborus as sister group to the RTA clade ldquocan bepresumed to be falserdquoMoreover molecular data analyses often fail to find
support for the monophyly of Deinopoideamdashthecribellate orb-weavers (Uloboridae + Deinopidae) (egDimitrov et al 2012 2013 Bond et al 2014 Fernan-dez et al 2014) In contrast the monophyly of Arane-oidea (the ecribellate orb-weavers) is well supported byboth morphological and molecular data but relation-ships among families remained unresolved for the mostpart (Hormiga and Griswold 2014 and referencestherein) until publication of two recent transcriptome-based phylogenetic analyses (Bond et al 2014Fernandez et al 2014)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 223
As the present study shows the long-held hypothesisof Orbiculariae monophyly continues to be overturnedby molecular data using both standard PCR-amplifiedgenetic markers (Dimitrov et al 2013) and more per-suasively transcriptomic data (Bond et al 2014Fernandez et al 2014) These recent studies place thecribellate orb-weavers (Deinopoidea which do notform a clade) with other groups rather than with theecribellate orb-weavers (Araneoidea) as the mono-phyly hypothesis demandsSpurious groupings in orbicularian analyses could
result from a number of well-known causes Missingdata have long been discussed with respect to theirpotential for affecting phylogenetic results (eg Kear-ney 2002 Wiens 2003 Wiens and Morrill 2011) Forthe cladistic problem discussed herein missing dataoccurred because of variable success in obtainingsequences for all markers and because of a certain lackof overlap across published analyses Sparse taxonsampling can also be a concern (eg Pollock et al2002 Hillis et al 2003) particularly at higher levelsbecause it may produce results that are difficult tointerpret in the absence of relevant higher taxa (eginsufficient representation of symphytognathoids inBlackledge et al 2009) or that are refuted with a den-ser taxon sample (eg in Lopardo and Hormiga 2008the addition of the family Synaphridae to the data ofGriswold et al 1998 changed the sister group ofCyatholipidae from Synotaxidae to Synaphridae)Another potential pitfall stems from unrecognized par-alogy (or lack of concerted evolution) of nuclear ribo-somal genes widely used in spider phylogenetic studiesNuclear rRNAs of some orbicularian spiders haveattracted attention because of their high variability notonly in total length but also at the nucleotide compo-sition level (eg Spagna and Gillespie 2006) Recentlya study specifically designed to test for paralogues ofthe 28S rRNA gene in jumping spiders found multiplecopies of this gene in a single specimen (Vink et al2011)Furthermore reconstructing the evolutionary chron-
icle of orb-weavers is a particularly onerous taskbecause araneoid family-level phylogeny is likely theresult of an ancient radiation compressed in a rela-tively narrow timespan (Dimitrov et al 2012) as hasalso been shown when reconstructing rapid radiationsof other major arthropod lineages such as in the lepi-dopteran phylogeny problem (eg Bazinet et al 2013)Published data (eg Dimitrov et al 2012 and refer-
ences therein) suggest a Late Triassic origin of orb-weavers and a late JurassicndashEarly Cretaceous originfor most araneoid families (but see Bond et al 2014for a proposed early Jurassic origin for the orb-web)The diversity of orbicularian species and lifestyles
including web architecture remains poorly understoodin part because of the lack of a robust phylogenetic
framework Standing questions include whether orb-webs were transformed into sheets cobwebs and otherforms (see Figs 6 and 7 for examples) multiple timesor if there was a single ldquolossrdquo of the typical orb archi-tecture defining a large clade of araneoids (for exam-ple as suggested in Griswold et al 1998) Of courseat shallow phylogenetic levels many such orb transfor-mations are known for example within Anapidaethere are transitions from orb- to sheet-webs Under-standing web evolution and diversification requires anempirically robust hypothesis about the underlyingphylogenetic patternsIn this study we have expanded the taxonomic sam-
ple used in our previous work (Dimitrov et al 2012)both within araneoids and their potential outgrouptaxa The main goal of this study is to test the limitsof Araneoidea using standard polymerase chain reac-tion (PCR)-amplified molecular markers and includingall current and former members of the superfamilyand to reconstruct the interfamilial relationships ofaraneoids In addition our analyses aim to provide aphylogenic framework with which to study web evolu-tion and diversification in araneoids and to set up aroadmap for future studies of araneoid relationshipsusing phylogenomic data
Materials and methods
Taxon sampling
The current study builds on the recent analyses ofDimitrov et al (2012) expanding greatly the taxonsampling of araneoid lineages with specific emphasison families and putative groups within families thatwere poorly represented or absent in former molecularphylogenies We have emphasized the addition of datafor families that were under-represented in our previ-ous study as well as those whose phylogenetic place-ment is critical to understand web evolution (eg inSynotaxidae synotaxine webs (ldquoregularrdquo Fig 6C) vspahorine physoglenine webs (ldquoirregularrdquo sheetsFig 7AndashF)) We also provide the first molecular datafor the araneoid family Synaphridae In addition anextended number of Palpimanoidea and other out-group taxa have been included in order to test the lim-its of Araneoidea and the controversial placement ofsome araneoid linages (eg Holarchaeidae) in Palpi-manoidea The present matrix thus brings together forthe first time representatives of all orbicularian fami-lies We have sequenced de novo 98 species and added265 species to the analyses using data from other stud-ies and those available in GenBank (Arnedo et al2007 2009 Rix et al 2008 Alvarez-Padilla et al2009 Blackledge et al 2009 Miller et al 2010 Dim-itrov and Hormiga 2011 Lopardo et al 2011
224 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Dimitrov et al 2012 Wood et al 2012) The com-plete list of taxa 363 terminals in total and theGenBank accession numbers are listed in Table S1Taxon names and nomenclatural changes are discussedin the ldquoSystematics of Araneoidea and Nicodamoideardquosection
Molecular methods
For each specimen up to three legs were used fortotal DNA extraction using the DNeasy tissue kit(Qiagen Valencia CA USA) the remainder of thespider was kept as a voucher Purified genomic DNAwas used as a template in order to target the followingsix genes or gene fragments two nuclear ribosomalgenes 18S rRNA (18S hereafter ~1800 bp) and 28SrRNA (28S hereafter fragment of ~2700 bp) twomitochondrial ribosomal genes 12S rRNA (12S here-after ~400 bp) and 16S rRNA (16S hereafter~550 bp) the nuclear protein-encoding gene histoneH3 (H3 hereafter 327 bp) and the mitochondrial pro-tein-encoding gene cytochrome c oxidase subunit I(COI hereafter 771 bp) We did not generate addi-tional wingless sequences as part of the current studyAll wingless sequences used in the analyses come fromprevious studies and were already available in Gen-Bank The PCRs were carried out using IllustraTMpuReTaq Ready-To-Go PCR beads (GE HealthcareUK wwwgelifesciencescom) as described in theSupporting InformationPCR-amplified products were sent to the High
Throughput Sequencing (htSEQ) Genomics Centerfacility at the University of Washington (Seattle WAUSA) for enzymatic cleanup and double-strandedsequencing The resulting chromatograms were readand edited and overlapping sequence fragments assem-bled visually inspected and edited using Sequencherv47 (Gene Codes Corporation Ann Harbor MIUSA) and Geneious v605 (Biomatters available athttpwwwgeneiouscom) In order to detect contam-ination individual fragments were submitted toBLAST (Basic Local Alignment Search Tool) asimplemented on the NCBI website (httpblastncbinlmnihgov) A consensus was compiledfrom all sequenced DNA fragments for each gene andtaxon and deposited in GenBank (Table S1) The bio-logical sequence alignment editor Bioedit v7111(Hall 1999 available at httpwwwmbioncsueduBioEditbioedithtml) was used to edit the completesequences
Phylogenetic analyses
All molecular phylogenetic analyses were run on theAbel Cluster at the University of Oslo the CIPRESscience gateway (Miller et al 2011) and at a Linux
server at the Natural History Museum Oslo Parsi-mony analyses were run on a fast desktop computer atthe Natural History Museum of Denmark Universityof Copenhagen
Alignments Multiple sequence alignments werecarried out with MAFFT v7058b (Katoh andStandley 2013) run on the Ubuntu server at theNatural History Museum University of OsloAlignments of protein-encoding genes were trivial dueto the lack of gaps (except few insertionsdeletions inwingless) and were produced using the L-INS-imethod Ribosomal genes however contain variableregions In addition the distribution of insertions anddeletions is nonrandom in stem regions due tostructural constraints such as compensatory mutationsand consequently taking rRNA secondary structureinto consideration is also important (Rix et al 2008Murienne et al 2010) To that end we have used theQ-INS-i method which implements the four-wayconsistency objective function (Katoh and Toh 2008)Because the Q-INS-i method is computationally verydemanding long fragments such as 18S and 28S werealigned in shorter blocks (based on amplicon limits)which were assembled after alignmentIn a few cases sequences were found to be a con-
tamination or potential paralogues and were excludedfrom the final analyses (see supporting information)However to exemplify the effect of indiscriminatelyincluding all data we ran a round of maximum-likeli-hood (ML) analyses keeping these sequences Theseresults are not discussed further here but are shown inFig S1 Additional data sets were created using differ-ent approaches to improve data completeness ordecrease potential ambiguities To increase data com-pleteness we excluded taxa that were not sequencedfor most of the genes in a stepwise fashion retainingtaxa with data for at least three genes and taxa withdata for at least four genes In order to reduceambiguously aligned regions in the data set we pro-cessed the ribosomal genes with the program trimalv13 (Capella-Gutierrez et al 2009) using the heuris-tic automated1 method and the gappyout method forthe 28S1 fragment for which automated1 failed to pro-vide plausible solution The list of all matrices and thetreatments that were applied to generate them aresummarized in Table S2
Maximum-likelihood The ML analyses were carriedout with the program RAxML (Stamatakis 2014) onCIPRES or on Abel The concatenated gene matrixwas partitioned by gene and the protein-encodinggenes were further partitioned into 1st + 2nd positionand 3rd position partitions Bootstrap and optimaltrees were computed in the same run using the faoption using 1000 bootstrapping replicates Trees were
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 225
rooted using the mygalomorph spider Euagruschisoseus (Dipluridae)
Nonparametric methods and mixture models Becauseeach position in a gene can be under different selectivepressures a site-specific approach to the estimation ofsubstitution rates and other model parameters may bemost appropriate To investigate the effects of thisapproximation we used the nonparametric models ofsite-specific rates of equilibrium frequency profiles asimplemented in PhyloBayes v33e (Lartillot et al2009) We used the CAT-GTR model which is themost appropriate for DNA (-cat -gtr -dgam 4) Twoindependent runs were launched and checked forconvergence and the results are summarized in thetopology presented in Fig S2
Parsimony methods The parsimony analyses of theconcatenated molecular matrix were carried out withthe computer program TNT v11 (Goloboff et al2008) Given the size of the matrix (363 taxa and 7genes) a driven search combining new technologyalgorithms using equal weights (ie tree drifting mixedsectorial searches and tree fusing) was performed (50initial addition sequences initial level 10 cycles ofdrifting 10) until it stabilized onto a strict consensusfive times (with default factor of 75) This is one of themost efficient search strategies when dealing withlarge difficult data sets (Goloboff 1999) Most othersearch settings were left as default values Commandsused were included in and run from a script filewhich was generated by modifying an automaticallygenerated TNT batch file The detailed sequence ofcommands is given in the Supporting InformationNodal support was estimated via 1000 replicates of
parsimony jackknifing (Farris et al 1996 Farris1997) under new technology (using default values)
Divergence time estimation In order to estimatedivergence times we used a relaxed uncorrelatedlognormal approximation (Drummond et al 2006) asimplemented in the program BEAST v211(Bouckaert et al 2014) Analyses in BEAST were runwith exponential distribution for the probabilitydensity of the tmrca prior and birthndashdeath model forthe tree prior Calibration points and relevant priorparameters are listed in Table S3 Parameters werechosen in such a way that 95 of the priorsrsquodistributions fell between the minimum (the offset) andthe maximum values reported for the datinguncertainty of the corresponding fossil Because it isunknown how far the fossil is from the most recentcommon ancestor of the node that it is constraining(eg what is its position along the stem) we used anoninformative hyper prior with gamma distribution toincorporate the uncertainty of the calibration-density
(Heath 2012) All constraints were applied as stemcalibrations In the results presented here we have notincluded as a constraint the fossil spiderMongolarachne jurassica (Selden et al 2011 2013formerly classified as a Nephila species) from theMiddle Jurassic deposits of China (Inner MongoliaDaohugou China) because of recent concerns aboutits taxonomic placement (eg Kuntner et al 2013)However the fossil described by Selden et al (2011)does seem to have morphological characters compatiblewith those of other nephilids A male specimendescribed two years later was assigned to the samespecies (Selden et al 2013) and because the male didnot fit the Nephilidae diagnosis the female (describedas N jurassica) and the male were placed in a newfamilymdashMongolarachnidae Selden et al (2013) didnot present convincing evidence that these twospecimens are conspecific (eg the male resemblesEctatosticta a hypochilid genus endemic to China) soin our view the question of where M jurassica belongsis still in need of further research For example recentdescription of Geratonephila burmanica from EarlyCretaceous Burmese amber (97ndash110 Myr old Poinarand Buckley 2012 see also Penney 2014) challengesthe hypothesis of Kuntner et al (2013) that the cladeof Nephila and its close relatives is only 40ndash60 Myr oldAs a starting tree in all BEAST runs we used the
best tree from the ML analysis of the full data set thatwas processed with the program treePL (Smith andOrsquoMeara 2012) and the same sets of calibration con-straints as for the corresponding BEAST analysesNodes where fossil calibrations were applied were alsoconstrained as monophyletic (note that these werealready selected in order to reflect well-supportedmonophyletic groups as found by the ML analysessee arrows on Fig 3) however the starting tree topol-ogy was not strictly constrained in order to accountfor topological uncertainties Conversion of the MLtree to ultrametric with treePL was necessary in orderto provide BEAST with a starting tree that satisfies allpriors and topological constraints Clock and substitu-tion models were unlinked between gene partitionsexcept for the mitochondrial genes (16S and COI)Analyses were run for at least 200 million generationswith second runs for at least 70 million generations totest for convergence of the results Chain mixing effec-tive sample sizes of estimates and other relevant statis-tics were evaluated in Tracer v15 (Rambaut andDrummond 2007) Trees were summarized with theprogram TreeAnnotator which is distributed as partof the BEAST package Two different sets of datinganalyses were run with calibrations applied in such away that the nephilids are treated as a clade with ara-neids (Araneidae) and as an independent clade (seediscussion in the ldquoSystematics of Araneoidea andNicodamoideardquo section) In addition to the partitioned
226 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
of radii and a sticky spiral (eg in TetragnathidaeFig 6F) to highly irregular tridimensional webs (egin Linyphiidae Fig 6D G H) Almost everythingin between these architectural extremes seems to existand most of this web diversity is still undiscoveredor undocumented (eg Scharff and Hormiga 2012)In some cases foraging webs have been abandonedaltogether such as in the pirate spiders (MimetidaeFig 4C)
Two groups of orb-weaversmdashdeinopoids and arane-oidsmdashbuild similar webs that differ significantly in thestructure and composition of the silk of their capturespiral Traditionally regarded as a lineage these twogroups are now hypothesized not to form a clade(Dimitrov et al 2013 Bond et al 2014 Fernandezet al 2014) Deinopoids (Deinopidae Uloboridae) usecribellate silk for their sticky spiral (Fig 1A B) whilethe allegedly homologous counterpart in araneoids is
(A) (B)
(C) (D)
Fig 1 (A) The cribellate web of Sybota sp (Uloboridae) from Chile (DSC_2250) (B) The cribellate ogre-face spider Deinopis sp (Deinopidae)from Australia (DSC_0983) (C) The ecribellate Nephila plumipes building its orb-web Australia the highly reflective silk lines in this image arethe viscid capture spiral turns covered with a sticky glycoprotein a synapomorphy of Araneoidea The less reflective silk lines in between stickyturns are part of the temporary nonsticky spiral which in Nephila and its relatives are left in the finished web (DSC_6451) (D) Progradungulaotwayensis (Gradungulidae) from Australia with its ladder cribellate web an example of an early-branching araneomorph that illustrates theantiquity of cribellate silk (DSC_1424) Photos G Hormiga
222 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
made of a type of viscid silk that is unique to arane-oids (eg Fig 1C) Cribellate silk is ancient (egFig 1D)mdashit evolved in the early araneomorph lin-eagesmdashand thus sharing such type of silk among dei-nopoid taxa is expected to be symplesiomorphic Thistype of silk is spun by a spinning plate (the cribellum)in combination with a combing structure on the fourthleg metatarsus consisting of a row of modifiedmacrosetae (the calamistrum) Cribellate silk is ldquodryrdquoand is formed of thousands of fine looped fibrilswoven on a core of two axial fibres (eg Opell 1998fig 1) Its adhesive properties are attained by van derWaals and hygroscopic forces (Hawthorn and Opell2003) In contrast araneoids produce a novel type ofsticky silk in which the axial fibres are coated with aviscid glycoprotein This type of composite stickythread is produced faster presumably more economi-cally and attains a much higher stickiness than thedry deinopoid cribellate silk A large body of empiricalwork has studied and compared the biological andphysicochemical properties of these types of silks (seereview in Blackledge 2012)There is a marked disparity in species richness
between cribellate and ecribellate orb-weavers Themajority of orb-weaving spiders are members of thesuperfamily Araneoidea (the ecribellate orb-weavers 17families more than 12 000 species described) In com-parison Deinopoidea the cribellate orb-weaversinclude only 331 described species in two families Nico-damidae a small Austral group (29 species named) withboth cribellate and ecribellate members appears to bephylogenetically related to the ecribellate orb-weavers(Blackledge et al 2009 Dimitrov et al 2012) Thisasymmetry in species diversity between deinopoids andaraneoids has been attributed to the shift in type of cap-ture thread from dry fuzzy cribellate silk (Fig 1B) toviscid sticky silk (Fig 1C) combined with changes inthe silk spectral reflective properties and a transitionfrom horizontal to vertical orb-webs (references summa-rized in Hormiga and Griswold 2014) However recentstudies (Dimitrov et al 2013 Bond et al 2014Fernandez et al 2014) and the results presented hereshow that the contrast DeinopoideandashAraneoidea is nolonger valid and it is likely that evolution of webs anddiversification into new ecological niches are responsiblefor the differences in diversity of these spider clades (egDimitrov et al 2012)The question of whether cribellate and ecribellate
orb-webs can be traced to a single origin or haveevolved independently began to be debated in the 19thCentury (summarized in Coddington 1986) and hasbeen discussed extensively in the literature It was notuntil the late 1980s that a consensus began to emergeon the answer to this problem During the last threedecades the combination of comparative behaviouraldata (such as the seminal work of Eberhard 1982) and
cladistic approaches to analyse the available evidencehas favoured a monophyletic origin of orb-webs andthe monophyly of Orbiculariae (eg Levi and Cod-dington 1983 Coddington 1986 1990) with the pre-ponderance of evidence supporting this view comingfrom the webs and the concomitant stereotypical beha-viours used to build them Most research in the lasttwo decades has supported a single origin of the orb-web Because the monophyly of orb-weavers has beensupported primarily by behavioural and spinningorgan characters it has been challenging to test thepossibility that orb-webs were not convergent in thecribellate and ecribellate orb-weavers without referringto the building behaviours and silk products Geneticdata have played an increasingly important role inresolving spider phylogenetic relationships mostly inthe form of nucleotide sequences from a few genes (thenuclear and mitochondrial rRNA genes 18S 28S 12Sand 16S and a handful of protein-encoding genes fromwhich the most commonly used are the nuclear histoneH3 and the mitochondrial COI) often humorouslydescribed as ldquothe usual suspectsrdquo However the suc-cess of these markers as an independent test to resolveorbicularian relationships has been limited particularlyat the interfamilial level (eg Blackledge et al 2009Dimitrov et al 2012)Only one phylogenetic analysis of molecular data
with a sufficiently dense taxon sample to properlyaddress interfamilial relationships has recovered Orbic-ulariae as a clade albeit without support (Dimitrovet al 2012) Furthermore these nucleotide data failedto resolve or provide support for the relationshipsamong most orbicularian families the majority of deepinternodes are short Although most phylogeneticanalyses of DNA sequence data have found that orbic-ularians are not monophyletic this particular resulthas often been dismissed as ldquoartefactualrdquo (eg due totaxon sampling effects) or ldquomisleadingrdquomdashsuch hasbeen the convincing power of the orbicularian mono-phyly hypothesis For example in an analysis of thespider sequences available in GenBank Agnarssonet al (2013) explicitly stated that the placement ofUloborus as sister group to the RTA clade ldquocan bepresumed to be falserdquoMoreover molecular data analyses often fail to find
support for the monophyly of Deinopoideamdashthecribellate orb-weavers (Uloboridae + Deinopidae) (egDimitrov et al 2012 2013 Bond et al 2014 Fernan-dez et al 2014) In contrast the monophyly of Arane-oidea (the ecribellate orb-weavers) is well supported byboth morphological and molecular data but relation-ships among families remained unresolved for the mostpart (Hormiga and Griswold 2014 and referencestherein) until publication of two recent transcriptome-based phylogenetic analyses (Bond et al 2014Fernandez et al 2014)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 223
As the present study shows the long-held hypothesisof Orbiculariae monophyly continues to be overturnedby molecular data using both standard PCR-amplifiedgenetic markers (Dimitrov et al 2013) and more per-suasively transcriptomic data (Bond et al 2014Fernandez et al 2014) These recent studies place thecribellate orb-weavers (Deinopoidea which do notform a clade) with other groups rather than with theecribellate orb-weavers (Araneoidea) as the mono-phyly hypothesis demandsSpurious groupings in orbicularian analyses could
result from a number of well-known causes Missingdata have long been discussed with respect to theirpotential for affecting phylogenetic results (eg Kear-ney 2002 Wiens 2003 Wiens and Morrill 2011) Forthe cladistic problem discussed herein missing dataoccurred because of variable success in obtainingsequences for all markers and because of a certain lackof overlap across published analyses Sparse taxonsampling can also be a concern (eg Pollock et al2002 Hillis et al 2003) particularly at higher levelsbecause it may produce results that are difficult tointerpret in the absence of relevant higher taxa (eginsufficient representation of symphytognathoids inBlackledge et al 2009) or that are refuted with a den-ser taxon sample (eg in Lopardo and Hormiga 2008the addition of the family Synaphridae to the data ofGriswold et al 1998 changed the sister group ofCyatholipidae from Synotaxidae to Synaphridae)Another potential pitfall stems from unrecognized par-alogy (or lack of concerted evolution) of nuclear ribo-somal genes widely used in spider phylogenetic studiesNuclear rRNAs of some orbicularian spiders haveattracted attention because of their high variability notonly in total length but also at the nucleotide compo-sition level (eg Spagna and Gillespie 2006) Recentlya study specifically designed to test for paralogues ofthe 28S rRNA gene in jumping spiders found multiplecopies of this gene in a single specimen (Vink et al2011)Furthermore reconstructing the evolutionary chron-
icle of orb-weavers is a particularly onerous taskbecause araneoid family-level phylogeny is likely theresult of an ancient radiation compressed in a rela-tively narrow timespan (Dimitrov et al 2012) as hasalso been shown when reconstructing rapid radiationsof other major arthropod lineages such as in the lepi-dopteran phylogeny problem (eg Bazinet et al 2013)Published data (eg Dimitrov et al 2012 and refer-
ences therein) suggest a Late Triassic origin of orb-weavers and a late JurassicndashEarly Cretaceous originfor most araneoid families (but see Bond et al 2014for a proposed early Jurassic origin for the orb-web)The diversity of orbicularian species and lifestyles
including web architecture remains poorly understoodin part because of the lack of a robust phylogenetic
framework Standing questions include whether orb-webs were transformed into sheets cobwebs and otherforms (see Figs 6 and 7 for examples) multiple timesor if there was a single ldquolossrdquo of the typical orb archi-tecture defining a large clade of araneoids (for exam-ple as suggested in Griswold et al 1998) Of courseat shallow phylogenetic levels many such orb transfor-mations are known for example within Anapidaethere are transitions from orb- to sheet-webs Under-standing web evolution and diversification requires anempirically robust hypothesis about the underlyingphylogenetic patternsIn this study we have expanded the taxonomic sam-
ple used in our previous work (Dimitrov et al 2012)both within araneoids and their potential outgrouptaxa The main goal of this study is to test the limitsof Araneoidea using standard polymerase chain reac-tion (PCR)-amplified molecular markers and includingall current and former members of the superfamilyand to reconstruct the interfamilial relationships ofaraneoids In addition our analyses aim to provide aphylogenic framework with which to study web evolu-tion and diversification in araneoids and to set up aroadmap for future studies of araneoid relationshipsusing phylogenomic data
Materials and methods
Taxon sampling
The current study builds on the recent analyses ofDimitrov et al (2012) expanding greatly the taxonsampling of araneoid lineages with specific emphasison families and putative groups within families thatwere poorly represented or absent in former molecularphylogenies We have emphasized the addition of datafor families that were under-represented in our previ-ous study as well as those whose phylogenetic place-ment is critical to understand web evolution (eg inSynotaxidae synotaxine webs (ldquoregularrdquo Fig 6C) vspahorine physoglenine webs (ldquoirregularrdquo sheetsFig 7AndashF)) We also provide the first molecular datafor the araneoid family Synaphridae In addition anextended number of Palpimanoidea and other out-group taxa have been included in order to test the lim-its of Araneoidea and the controversial placement ofsome araneoid linages (eg Holarchaeidae) in Palpi-manoidea The present matrix thus brings together forthe first time representatives of all orbicularian fami-lies We have sequenced de novo 98 species and added265 species to the analyses using data from other stud-ies and those available in GenBank (Arnedo et al2007 2009 Rix et al 2008 Alvarez-Padilla et al2009 Blackledge et al 2009 Miller et al 2010 Dim-itrov and Hormiga 2011 Lopardo et al 2011
224 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Dimitrov et al 2012 Wood et al 2012) The com-plete list of taxa 363 terminals in total and theGenBank accession numbers are listed in Table S1Taxon names and nomenclatural changes are discussedin the ldquoSystematics of Araneoidea and Nicodamoideardquosection
Molecular methods
For each specimen up to three legs were used fortotal DNA extraction using the DNeasy tissue kit(Qiagen Valencia CA USA) the remainder of thespider was kept as a voucher Purified genomic DNAwas used as a template in order to target the followingsix genes or gene fragments two nuclear ribosomalgenes 18S rRNA (18S hereafter ~1800 bp) and 28SrRNA (28S hereafter fragment of ~2700 bp) twomitochondrial ribosomal genes 12S rRNA (12S here-after ~400 bp) and 16S rRNA (16S hereafter~550 bp) the nuclear protein-encoding gene histoneH3 (H3 hereafter 327 bp) and the mitochondrial pro-tein-encoding gene cytochrome c oxidase subunit I(COI hereafter 771 bp) We did not generate addi-tional wingless sequences as part of the current studyAll wingless sequences used in the analyses come fromprevious studies and were already available in Gen-Bank The PCRs were carried out using IllustraTMpuReTaq Ready-To-Go PCR beads (GE HealthcareUK wwwgelifesciencescom) as described in theSupporting InformationPCR-amplified products were sent to the High
Throughput Sequencing (htSEQ) Genomics Centerfacility at the University of Washington (Seattle WAUSA) for enzymatic cleanup and double-strandedsequencing The resulting chromatograms were readand edited and overlapping sequence fragments assem-bled visually inspected and edited using Sequencherv47 (Gene Codes Corporation Ann Harbor MIUSA) and Geneious v605 (Biomatters available athttpwwwgeneiouscom) In order to detect contam-ination individual fragments were submitted toBLAST (Basic Local Alignment Search Tool) asimplemented on the NCBI website (httpblastncbinlmnihgov) A consensus was compiledfrom all sequenced DNA fragments for each gene andtaxon and deposited in GenBank (Table S1) The bio-logical sequence alignment editor Bioedit v7111(Hall 1999 available at httpwwwmbioncsueduBioEditbioedithtml) was used to edit the completesequences
Phylogenetic analyses
All molecular phylogenetic analyses were run on theAbel Cluster at the University of Oslo the CIPRESscience gateway (Miller et al 2011) and at a Linux
server at the Natural History Museum Oslo Parsi-mony analyses were run on a fast desktop computer atthe Natural History Museum of Denmark Universityof Copenhagen
Alignments Multiple sequence alignments werecarried out with MAFFT v7058b (Katoh andStandley 2013) run on the Ubuntu server at theNatural History Museum University of OsloAlignments of protein-encoding genes were trivial dueto the lack of gaps (except few insertionsdeletions inwingless) and were produced using the L-INS-imethod Ribosomal genes however contain variableregions In addition the distribution of insertions anddeletions is nonrandom in stem regions due tostructural constraints such as compensatory mutationsand consequently taking rRNA secondary structureinto consideration is also important (Rix et al 2008Murienne et al 2010) To that end we have used theQ-INS-i method which implements the four-wayconsistency objective function (Katoh and Toh 2008)Because the Q-INS-i method is computationally verydemanding long fragments such as 18S and 28S werealigned in shorter blocks (based on amplicon limits)which were assembled after alignmentIn a few cases sequences were found to be a con-
tamination or potential paralogues and were excludedfrom the final analyses (see supporting information)However to exemplify the effect of indiscriminatelyincluding all data we ran a round of maximum-likeli-hood (ML) analyses keeping these sequences Theseresults are not discussed further here but are shown inFig S1 Additional data sets were created using differ-ent approaches to improve data completeness ordecrease potential ambiguities To increase data com-pleteness we excluded taxa that were not sequencedfor most of the genes in a stepwise fashion retainingtaxa with data for at least three genes and taxa withdata for at least four genes In order to reduceambiguously aligned regions in the data set we pro-cessed the ribosomal genes with the program trimalv13 (Capella-Gutierrez et al 2009) using the heuris-tic automated1 method and the gappyout method forthe 28S1 fragment for which automated1 failed to pro-vide plausible solution The list of all matrices and thetreatments that were applied to generate them aresummarized in Table S2
Maximum-likelihood The ML analyses were carriedout with the program RAxML (Stamatakis 2014) onCIPRES or on Abel The concatenated gene matrixwas partitioned by gene and the protein-encodinggenes were further partitioned into 1st + 2nd positionand 3rd position partitions Bootstrap and optimaltrees were computed in the same run using the faoption using 1000 bootstrapping replicates Trees were
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 225
rooted using the mygalomorph spider Euagruschisoseus (Dipluridae)
Nonparametric methods and mixture models Becauseeach position in a gene can be under different selectivepressures a site-specific approach to the estimation ofsubstitution rates and other model parameters may bemost appropriate To investigate the effects of thisapproximation we used the nonparametric models ofsite-specific rates of equilibrium frequency profiles asimplemented in PhyloBayes v33e (Lartillot et al2009) We used the CAT-GTR model which is themost appropriate for DNA (-cat -gtr -dgam 4) Twoindependent runs were launched and checked forconvergence and the results are summarized in thetopology presented in Fig S2
Parsimony methods The parsimony analyses of theconcatenated molecular matrix were carried out withthe computer program TNT v11 (Goloboff et al2008) Given the size of the matrix (363 taxa and 7genes) a driven search combining new technologyalgorithms using equal weights (ie tree drifting mixedsectorial searches and tree fusing) was performed (50initial addition sequences initial level 10 cycles ofdrifting 10) until it stabilized onto a strict consensusfive times (with default factor of 75) This is one of themost efficient search strategies when dealing withlarge difficult data sets (Goloboff 1999) Most othersearch settings were left as default values Commandsused were included in and run from a script filewhich was generated by modifying an automaticallygenerated TNT batch file The detailed sequence ofcommands is given in the Supporting InformationNodal support was estimated via 1000 replicates of
parsimony jackknifing (Farris et al 1996 Farris1997) under new technology (using default values)
Divergence time estimation In order to estimatedivergence times we used a relaxed uncorrelatedlognormal approximation (Drummond et al 2006) asimplemented in the program BEAST v211(Bouckaert et al 2014) Analyses in BEAST were runwith exponential distribution for the probabilitydensity of the tmrca prior and birthndashdeath model forthe tree prior Calibration points and relevant priorparameters are listed in Table S3 Parameters werechosen in such a way that 95 of the priorsrsquodistributions fell between the minimum (the offset) andthe maximum values reported for the datinguncertainty of the corresponding fossil Because it isunknown how far the fossil is from the most recentcommon ancestor of the node that it is constraining(eg what is its position along the stem) we used anoninformative hyper prior with gamma distribution toincorporate the uncertainty of the calibration-density
(Heath 2012) All constraints were applied as stemcalibrations In the results presented here we have notincluded as a constraint the fossil spiderMongolarachne jurassica (Selden et al 2011 2013formerly classified as a Nephila species) from theMiddle Jurassic deposits of China (Inner MongoliaDaohugou China) because of recent concerns aboutits taxonomic placement (eg Kuntner et al 2013)However the fossil described by Selden et al (2011)does seem to have morphological characters compatiblewith those of other nephilids A male specimendescribed two years later was assigned to the samespecies (Selden et al 2013) and because the male didnot fit the Nephilidae diagnosis the female (describedas N jurassica) and the male were placed in a newfamilymdashMongolarachnidae Selden et al (2013) didnot present convincing evidence that these twospecimens are conspecific (eg the male resemblesEctatosticta a hypochilid genus endemic to China) soin our view the question of where M jurassica belongsis still in need of further research For example recentdescription of Geratonephila burmanica from EarlyCretaceous Burmese amber (97ndash110 Myr old Poinarand Buckley 2012 see also Penney 2014) challengesthe hypothesis of Kuntner et al (2013) that the cladeof Nephila and its close relatives is only 40ndash60 Myr oldAs a starting tree in all BEAST runs we used the
best tree from the ML analysis of the full data set thatwas processed with the program treePL (Smith andOrsquoMeara 2012) and the same sets of calibration con-straints as for the corresponding BEAST analysesNodes where fossil calibrations were applied were alsoconstrained as monophyletic (note that these werealready selected in order to reflect well-supportedmonophyletic groups as found by the ML analysessee arrows on Fig 3) however the starting tree topol-ogy was not strictly constrained in order to accountfor topological uncertainties Conversion of the MLtree to ultrametric with treePL was necessary in orderto provide BEAST with a starting tree that satisfies allpriors and topological constraints Clock and substitu-tion models were unlinked between gene partitionsexcept for the mitochondrial genes (16S and COI)Analyses were run for at least 200 million generationswith second runs for at least 70 million generations totest for convergence of the results Chain mixing effec-tive sample sizes of estimates and other relevant statis-tics were evaluated in Tracer v15 (Rambaut andDrummond 2007) Trees were summarized with theprogram TreeAnnotator which is distributed as partof the BEAST package Two different sets of datinganalyses were run with calibrations applied in such away that the nephilids are treated as a clade with ara-neids (Araneidae) and as an independent clade (seediscussion in the ldquoSystematics of Araneoidea andNicodamoideardquo section) In addition to the partitioned
226 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
made of a type of viscid silk that is unique to arane-oids (eg Fig 1C) Cribellate silk is ancient (egFig 1D)mdashit evolved in the early araneomorph lin-eagesmdashand thus sharing such type of silk among dei-nopoid taxa is expected to be symplesiomorphic Thistype of silk is spun by a spinning plate (the cribellum)in combination with a combing structure on the fourthleg metatarsus consisting of a row of modifiedmacrosetae (the calamistrum) Cribellate silk is ldquodryrdquoand is formed of thousands of fine looped fibrilswoven on a core of two axial fibres (eg Opell 1998fig 1) Its adhesive properties are attained by van derWaals and hygroscopic forces (Hawthorn and Opell2003) In contrast araneoids produce a novel type ofsticky silk in which the axial fibres are coated with aviscid glycoprotein This type of composite stickythread is produced faster presumably more economi-cally and attains a much higher stickiness than thedry deinopoid cribellate silk A large body of empiricalwork has studied and compared the biological andphysicochemical properties of these types of silks (seereview in Blackledge 2012)There is a marked disparity in species richness
between cribellate and ecribellate orb-weavers Themajority of orb-weaving spiders are members of thesuperfamily Araneoidea (the ecribellate orb-weavers 17families more than 12 000 species described) In com-parison Deinopoidea the cribellate orb-weaversinclude only 331 described species in two families Nico-damidae a small Austral group (29 species named) withboth cribellate and ecribellate members appears to bephylogenetically related to the ecribellate orb-weavers(Blackledge et al 2009 Dimitrov et al 2012) Thisasymmetry in species diversity between deinopoids andaraneoids has been attributed to the shift in type of cap-ture thread from dry fuzzy cribellate silk (Fig 1B) toviscid sticky silk (Fig 1C) combined with changes inthe silk spectral reflective properties and a transitionfrom horizontal to vertical orb-webs (references summa-rized in Hormiga and Griswold 2014) However recentstudies (Dimitrov et al 2013 Bond et al 2014Fernandez et al 2014) and the results presented hereshow that the contrast DeinopoideandashAraneoidea is nolonger valid and it is likely that evolution of webs anddiversification into new ecological niches are responsiblefor the differences in diversity of these spider clades (egDimitrov et al 2012)The question of whether cribellate and ecribellate
orb-webs can be traced to a single origin or haveevolved independently began to be debated in the 19thCentury (summarized in Coddington 1986) and hasbeen discussed extensively in the literature It was notuntil the late 1980s that a consensus began to emergeon the answer to this problem During the last threedecades the combination of comparative behaviouraldata (such as the seminal work of Eberhard 1982) and
cladistic approaches to analyse the available evidencehas favoured a monophyletic origin of orb-webs andthe monophyly of Orbiculariae (eg Levi and Cod-dington 1983 Coddington 1986 1990) with the pre-ponderance of evidence supporting this view comingfrom the webs and the concomitant stereotypical beha-viours used to build them Most research in the lasttwo decades has supported a single origin of the orb-web Because the monophyly of orb-weavers has beensupported primarily by behavioural and spinningorgan characters it has been challenging to test thepossibility that orb-webs were not convergent in thecribellate and ecribellate orb-weavers without referringto the building behaviours and silk products Geneticdata have played an increasingly important role inresolving spider phylogenetic relationships mostly inthe form of nucleotide sequences from a few genes (thenuclear and mitochondrial rRNA genes 18S 28S 12Sand 16S and a handful of protein-encoding genes fromwhich the most commonly used are the nuclear histoneH3 and the mitochondrial COI) often humorouslydescribed as ldquothe usual suspectsrdquo However the suc-cess of these markers as an independent test to resolveorbicularian relationships has been limited particularlyat the interfamilial level (eg Blackledge et al 2009Dimitrov et al 2012)Only one phylogenetic analysis of molecular data
with a sufficiently dense taxon sample to properlyaddress interfamilial relationships has recovered Orbic-ulariae as a clade albeit without support (Dimitrovet al 2012) Furthermore these nucleotide data failedto resolve or provide support for the relationshipsamong most orbicularian families the majority of deepinternodes are short Although most phylogeneticanalyses of DNA sequence data have found that orbic-ularians are not monophyletic this particular resulthas often been dismissed as ldquoartefactualrdquo (eg due totaxon sampling effects) or ldquomisleadingrdquomdashsuch hasbeen the convincing power of the orbicularian mono-phyly hypothesis For example in an analysis of thespider sequences available in GenBank Agnarssonet al (2013) explicitly stated that the placement ofUloborus as sister group to the RTA clade ldquocan bepresumed to be falserdquoMoreover molecular data analyses often fail to find
support for the monophyly of Deinopoideamdashthecribellate orb-weavers (Uloboridae + Deinopidae) (egDimitrov et al 2012 2013 Bond et al 2014 Fernan-dez et al 2014) In contrast the monophyly of Arane-oidea (the ecribellate orb-weavers) is well supported byboth morphological and molecular data but relation-ships among families remained unresolved for the mostpart (Hormiga and Griswold 2014 and referencestherein) until publication of two recent transcriptome-based phylogenetic analyses (Bond et al 2014Fernandez et al 2014)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 223
As the present study shows the long-held hypothesisof Orbiculariae monophyly continues to be overturnedby molecular data using both standard PCR-amplifiedgenetic markers (Dimitrov et al 2013) and more per-suasively transcriptomic data (Bond et al 2014Fernandez et al 2014) These recent studies place thecribellate orb-weavers (Deinopoidea which do notform a clade) with other groups rather than with theecribellate orb-weavers (Araneoidea) as the mono-phyly hypothesis demandsSpurious groupings in orbicularian analyses could
result from a number of well-known causes Missingdata have long been discussed with respect to theirpotential for affecting phylogenetic results (eg Kear-ney 2002 Wiens 2003 Wiens and Morrill 2011) Forthe cladistic problem discussed herein missing dataoccurred because of variable success in obtainingsequences for all markers and because of a certain lackof overlap across published analyses Sparse taxonsampling can also be a concern (eg Pollock et al2002 Hillis et al 2003) particularly at higher levelsbecause it may produce results that are difficult tointerpret in the absence of relevant higher taxa (eginsufficient representation of symphytognathoids inBlackledge et al 2009) or that are refuted with a den-ser taxon sample (eg in Lopardo and Hormiga 2008the addition of the family Synaphridae to the data ofGriswold et al 1998 changed the sister group ofCyatholipidae from Synotaxidae to Synaphridae)Another potential pitfall stems from unrecognized par-alogy (or lack of concerted evolution) of nuclear ribo-somal genes widely used in spider phylogenetic studiesNuclear rRNAs of some orbicularian spiders haveattracted attention because of their high variability notonly in total length but also at the nucleotide compo-sition level (eg Spagna and Gillespie 2006) Recentlya study specifically designed to test for paralogues ofthe 28S rRNA gene in jumping spiders found multiplecopies of this gene in a single specimen (Vink et al2011)Furthermore reconstructing the evolutionary chron-
icle of orb-weavers is a particularly onerous taskbecause araneoid family-level phylogeny is likely theresult of an ancient radiation compressed in a rela-tively narrow timespan (Dimitrov et al 2012) as hasalso been shown when reconstructing rapid radiationsof other major arthropod lineages such as in the lepi-dopteran phylogeny problem (eg Bazinet et al 2013)Published data (eg Dimitrov et al 2012 and refer-
ences therein) suggest a Late Triassic origin of orb-weavers and a late JurassicndashEarly Cretaceous originfor most araneoid families (but see Bond et al 2014for a proposed early Jurassic origin for the orb-web)The diversity of orbicularian species and lifestyles
including web architecture remains poorly understoodin part because of the lack of a robust phylogenetic
framework Standing questions include whether orb-webs were transformed into sheets cobwebs and otherforms (see Figs 6 and 7 for examples) multiple timesor if there was a single ldquolossrdquo of the typical orb archi-tecture defining a large clade of araneoids (for exam-ple as suggested in Griswold et al 1998) Of courseat shallow phylogenetic levels many such orb transfor-mations are known for example within Anapidaethere are transitions from orb- to sheet-webs Under-standing web evolution and diversification requires anempirically robust hypothesis about the underlyingphylogenetic patternsIn this study we have expanded the taxonomic sam-
ple used in our previous work (Dimitrov et al 2012)both within araneoids and their potential outgrouptaxa The main goal of this study is to test the limitsof Araneoidea using standard polymerase chain reac-tion (PCR)-amplified molecular markers and includingall current and former members of the superfamilyand to reconstruct the interfamilial relationships ofaraneoids In addition our analyses aim to provide aphylogenic framework with which to study web evolu-tion and diversification in araneoids and to set up aroadmap for future studies of araneoid relationshipsusing phylogenomic data
Materials and methods
Taxon sampling
The current study builds on the recent analyses ofDimitrov et al (2012) expanding greatly the taxonsampling of araneoid lineages with specific emphasison families and putative groups within families thatwere poorly represented or absent in former molecularphylogenies We have emphasized the addition of datafor families that were under-represented in our previ-ous study as well as those whose phylogenetic place-ment is critical to understand web evolution (eg inSynotaxidae synotaxine webs (ldquoregularrdquo Fig 6C) vspahorine physoglenine webs (ldquoirregularrdquo sheetsFig 7AndashF)) We also provide the first molecular datafor the araneoid family Synaphridae In addition anextended number of Palpimanoidea and other out-group taxa have been included in order to test the lim-its of Araneoidea and the controversial placement ofsome araneoid linages (eg Holarchaeidae) in Palpi-manoidea The present matrix thus brings together forthe first time representatives of all orbicularian fami-lies We have sequenced de novo 98 species and added265 species to the analyses using data from other stud-ies and those available in GenBank (Arnedo et al2007 2009 Rix et al 2008 Alvarez-Padilla et al2009 Blackledge et al 2009 Miller et al 2010 Dim-itrov and Hormiga 2011 Lopardo et al 2011
224 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Dimitrov et al 2012 Wood et al 2012) The com-plete list of taxa 363 terminals in total and theGenBank accession numbers are listed in Table S1Taxon names and nomenclatural changes are discussedin the ldquoSystematics of Araneoidea and Nicodamoideardquosection
Molecular methods
For each specimen up to three legs were used fortotal DNA extraction using the DNeasy tissue kit(Qiagen Valencia CA USA) the remainder of thespider was kept as a voucher Purified genomic DNAwas used as a template in order to target the followingsix genes or gene fragments two nuclear ribosomalgenes 18S rRNA (18S hereafter ~1800 bp) and 28SrRNA (28S hereafter fragment of ~2700 bp) twomitochondrial ribosomal genes 12S rRNA (12S here-after ~400 bp) and 16S rRNA (16S hereafter~550 bp) the nuclear protein-encoding gene histoneH3 (H3 hereafter 327 bp) and the mitochondrial pro-tein-encoding gene cytochrome c oxidase subunit I(COI hereafter 771 bp) We did not generate addi-tional wingless sequences as part of the current studyAll wingless sequences used in the analyses come fromprevious studies and were already available in Gen-Bank The PCRs were carried out using IllustraTMpuReTaq Ready-To-Go PCR beads (GE HealthcareUK wwwgelifesciencescom) as described in theSupporting InformationPCR-amplified products were sent to the High
Throughput Sequencing (htSEQ) Genomics Centerfacility at the University of Washington (Seattle WAUSA) for enzymatic cleanup and double-strandedsequencing The resulting chromatograms were readand edited and overlapping sequence fragments assem-bled visually inspected and edited using Sequencherv47 (Gene Codes Corporation Ann Harbor MIUSA) and Geneious v605 (Biomatters available athttpwwwgeneiouscom) In order to detect contam-ination individual fragments were submitted toBLAST (Basic Local Alignment Search Tool) asimplemented on the NCBI website (httpblastncbinlmnihgov) A consensus was compiledfrom all sequenced DNA fragments for each gene andtaxon and deposited in GenBank (Table S1) The bio-logical sequence alignment editor Bioedit v7111(Hall 1999 available at httpwwwmbioncsueduBioEditbioedithtml) was used to edit the completesequences
Phylogenetic analyses
All molecular phylogenetic analyses were run on theAbel Cluster at the University of Oslo the CIPRESscience gateway (Miller et al 2011) and at a Linux
server at the Natural History Museum Oslo Parsi-mony analyses were run on a fast desktop computer atthe Natural History Museum of Denmark Universityof Copenhagen
Alignments Multiple sequence alignments werecarried out with MAFFT v7058b (Katoh andStandley 2013) run on the Ubuntu server at theNatural History Museum University of OsloAlignments of protein-encoding genes were trivial dueto the lack of gaps (except few insertionsdeletions inwingless) and were produced using the L-INS-imethod Ribosomal genes however contain variableregions In addition the distribution of insertions anddeletions is nonrandom in stem regions due tostructural constraints such as compensatory mutationsand consequently taking rRNA secondary structureinto consideration is also important (Rix et al 2008Murienne et al 2010) To that end we have used theQ-INS-i method which implements the four-wayconsistency objective function (Katoh and Toh 2008)Because the Q-INS-i method is computationally verydemanding long fragments such as 18S and 28S werealigned in shorter blocks (based on amplicon limits)which were assembled after alignmentIn a few cases sequences were found to be a con-
tamination or potential paralogues and were excludedfrom the final analyses (see supporting information)However to exemplify the effect of indiscriminatelyincluding all data we ran a round of maximum-likeli-hood (ML) analyses keeping these sequences Theseresults are not discussed further here but are shown inFig S1 Additional data sets were created using differ-ent approaches to improve data completeness ordecrease potential ambiguities To increase data com-pleteness we excluded taxa that were not sequencedfor most of the genes in a stepwise fashion retainingtaxa with data for at least three genes and taxa withdata for at least four genes In order to reduceambiguously aligned regions in the data set we pro-cessed the ribosomal genes with the program trimalv13 (Capella-Gutierrez et al 2009) using the heuris-tic automated1 method and the gappyout method forthe 28S1 fragment for which automated1 failed to pro-vide plausible solution The list of all matrices and thetreatments that were applied to generate them aresummarized in Table S2
Maximum-likelihood The ML analyses were carriedout with the program RAxML (Stamatakis 2014) onCIPRES or on Abel The concatenated gene matrixwas partitioned by gene and the protein-encodinggenes were further partitioned into 1st + 2nd positionand 3rd position partitions Bootstrap and optimaltrees were computed in the same run using the faoption using 1000 bootstrapping replicates Trees were
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 225
rooted using the mygalomorph spider Euagruschisoseus (Dipluridae)
Nonparametric methods and mixture models Becauseeach position in a gene can be under different selectivepressures a site-specific approach to the estimation ofsubstitution rates and other model parameters may bemost appropriate To investigate the effects of thisapproximation we used the nonparametric models ofsite-specific rates of equilibrium frequency profiles asimplemented in PhyloBayes v33e (Lartillot et al2009) We used the CAT-GTR model which is themost appropriate for DNA (-cat -gtr -dgam 4) Twoindependent runs were launched and checked forconvergence and the results are summarized in thetopology presented in Fig S2
Parsimony methods The parsimony analyses of theconcatenated molecular matrix were carried out withthe computer program TNT v11 (Goloboff et al2008) Given the size of the matrix (363 taxa and 7genes) a driven search combining new technologyalgorithms using equal weights (ie tree drifting mixedsectorial searches and tree fusing) was performed (50initial addition sequences initial level 10 cycles ofdrifting 10) until it stabilized onto a strict consensusfive times (with default factor of 75) This is one of themost efficient search strategies when dealing withlarge difficult data sets (Goloboff 1999) Most othersearch settings were left as default values Commandsused were included in and run from a script filewhich was generated by modifying an automaticallygenerated TNT batch file The detailed sequence ofcommands is given in the Supporting InformationNodal support was estimated via 1000 replicates of
parsimony jackknifing (Farris et al 1996 Farris1997) under new technology (using default values)
Divergence time estimation In order to estimatedivergence times we used a relaxed uncorrelatedlognormal approximation (Drummond et al 2006) asimplemented in the program BEAST v211(Bouckaert et al 2014) Analyses in BEAST were runwith exponential distribution for the probabilitydensity of the tmrca prior and birthndashdeath model forthe tree prior Calibration points and relevant priorparameters are listed in Table S3 Parameters werechosen in such a way that 95 of the priorsrsquodistributions fell between the minimum (the offset) andthe maximum values reported for the datinguncertainty of the corresponding fossil Because it isunknown how far the fossil is from the most recentcommon ancestor of the node that it is constraining(eg what is its position along the stem) we used anoninformative hyper prior with gamma distribution toincorporate the uncertainty of the calibration-density
(Heath 2012) All constraints were applied as stemcalibrations In the results presented here we have notincluded as a constraint the fossil spiderMongolarachne jurassica (Selden et al 2011 2013formerly classified as a Nephila species) from theMiddle Jurassic deposits of China (Inner MongoliaDaohugou China) because of recent concerns aboutits taxonomic placement (eg Kuntner et al 2013)However the fossil described by Selden et al (2011)does seem to have morphological characters compatiblewith those of other nephilids A male specimendescribed two years later was assigned to the samespecies (Selden et al 2013) and because the male didnot fit the Nephilidae diagnosis the female (describedas N jurassica) and the male were placed in a newfamilymdashMongolarachnidae Selden et al (2013) didnot present convincing evidence that these twospecimens are conspecific (eg the male resemblesEctatosticta a hypochilid genus endemic to China) soin our view the question of where M jurassica belongsis still in need of further research For example recentdescription of Geratonephila burmanica from EarlyCretaceous Burmese amber (97ndash110 Myr old Poinarand Buckley 2012 see also Penney 2014) challengesthe hypothesis of Kuntner et al (2013) that the cladeof Nephila and its close relatives is only 40ndash60 Myr oldAs a starting tree in all BEAST runs we used the
best tree from the ML analysis of the full data set thatwas processed with the program treePL (Smith andOrsquoMeara 2012) and the same sets of calibration con-straints as for the corresponding BEAST analysesNodes where fossil calibrations were applied were alsoconstrained as monophyletic (note that these werealready selected in order to reflect well-supportedmonophyletic groups as found by the ML analysessee arrows on Fig 3) however the starting tree topol-ogy was not strictly constrained in order to accountfor topological uncertainties Conversion of the MLtree to ultrametric with treePL was necessary in orderto provide BEAST with a starting tree that satisfies allpriors and topological constraints Clock and substitu-tion models were unlinked between gene partitionsexcept for the mitochondrial genes (16S and COI)Analyses were run for at least 200 million generationswith second runs for at least 70 million generations totest for convergence of the results Chain mixing effec-tive sample sizes of estimates and other relevant statis-tics were evaluated in Tracer v15 (Rambaut andDrummond 2007) Trees were summarized with theprogram TreeAnnotator which is distributed as partof the BEAST package Two different sets of datinganalyses were run with calibrations applied in such away that the nephilids are treated as a clade with ara-neids (Araneidae) and as an independent clade (seediscussion in the ldquoSystematics of Araneoidea andNicodamoideardquo section) In addition to the partitioned
226 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
As the present study shows the long-held hypothesisof Orbiculariae monophyly continues to be overturnedby molecular data using both standard PCR-amplifiedgenetic markers (Dimitrov et al 2013) and more per-suasively transcriptomic data (Bond et al 2014Fernandez et al 2014) These recent studies place thecribellate orb-weavers (Deinopoidea which do notform a clade) with other groups rather than with theecribellate orb-weavers (Araneoidea) as the mono-phyly hypothesis demandsSpurious groupings in orbicularian analyses could
result from a number of well-known causes Missingdata have long been discussed with respect to theirpotential for affecting phylogenetic results (eg Kear-ney 2002 Wiens 2003 Wiens and Morrill 2011) Forthe cladistic problem discussed herein missing dataoccurred because of variable success in obtainingsequences for all markers and because of a certain lackof overlap across published analyses Sparse taxonsampling can also be a concern (eg Pollock et al2002 Hillis et al 2003) particularly at higher levelsbecause it may produce results that are difficult tointerpret in the absence of relevant higher taxa (eginsufficient representation of symphytognathoids inBlackledge et al 2009) or that are refuted with a den-ser taxon sample (eg in Lopardo and Hormiga 2008the addition of the family Synaphridae to the data ofGriswold et al 1998 changed the sister group ofCyatholipidae from Synotaxidae to Synaphridae)Another potential pitfall stems from unrecognized par-alogy (or lack of concerted evolution) of nuclear ribo-somal genes widely used in spider phylogenetic studiesNuclear rRNAs of some orbicularian spiders haveattracted attention because of their high variability notonly in total length but also at the nucleotide compo-sition level (eg Spagna and Gillespie 2006) Recentlya study specifically designed to test for paralogues ofthe 28S rRNA gene in jumping spiders found multiplecopies of this gene in a single specimen (Vink et al2011)Furthermore reconstructing the evolutionary chron-
icle of orb-weavers is a particularly onerous taskbecause araneoid family-level phylogeny is likely theresult of an ancient radiation compressed in a rela-tively narrow timespan (Dimitrov et al 2012) as hasalso been shown when reconstructing rapid radiationsof other major arthropod lineages such as in the lepi-dopteran phylogeny problem (eg Bazinet et al 2013)Published data (eg Dimitrov et al 2012 and refer-
ences therein) suggest a Late Triassic origin of orb-weavers and a late JurassicndashEarly Cretaceous originfor most araneoid families (but see Bond et al 2014for a proposed early Jurassic origin for the orb-web)The diversity of orbicularian species and lifestyles
including web architecture remains poorly understoodin part because of the lack of a robust phylogenetic
framework Standing questions include whether orb-webs were transformed into sheets cobwebs and otherforms (see Figs 6 and 7 for examples) multiple timesor if there was a single ldquolossrdquo of the typical orb archi-tecture defining a large clade of araneoids (for exam-ple as suggested in Griswold et al 1998) Of courseat shallow phylogenetic levels many such orb transfor-mations are known for example within Anapidaethere are transitions from orb- to sheet-webs Under-standing web evolution and diversification requires anempirically robust hypothesis about the underlyingphylogenetic patternsIn this study we have expanded the taxonomic sam-
ple used in our previous work (Dimitrov et al 2012)both within araneoids and their potential outgrouptaxa The main goal of this study is to test the limitsof Araneoidea using standard polymerase chain reac-tion (PCR)-amplified molecular markers and includingall current and former members of the superfamilyand to reconstruct the interfamilial relationships ofaraneoids In addition our analyses aim to provide aphylogenic framework with which to study web evolu-tion and diversification in araneoids and to set up aroadmap for future studies of araneoid relationshipsusing phylogenomic data
Materials and methods
Taxon sampling
The current study builds on the recent analyses ofDimitrov et al (2012) expanding greatly the taxonsampling of araneoid lineages with specific emphasison families and putative groups within families thatwere poorly represented or absent in former molecularphylogenies We have emphasized the addition of datafor families that were under-represented in our previ-ous study as well as those whose phylogenetic place-ment is critical to understand web evolution (eg inSynotaxidae synotaxine webs (ldquoregularrdquo Fig 6C) vspahorine physoglenine webs (ldquoirregularrdquo sheetsFig 7AndashF)) We also provide the first molecular datafor the araneoid family Synaphridae In addition anextended number of Palpimanoidea and other out-group taxa have been included in order to test the lim-its of Araneoidea and the controversial placement ofsome araneoid linages (eg Holarchaeidae) in Palpi-manoidea The present matrix thus brings together forthe first time representatives of all orbicularian fami-lies We have sequenced de novo 98 species and added265 species to the analyses using data from other stud-ies and those available in GenBank (Arnedo et al2007 2009 Rix et al 2008 Alvarez-Padilla et al2009 Blackledge et al 2009 Miller et al 2010 Dim-itrov and Hormiga 2011 Lopardo et al 2011
224 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Dimitrov et al 2012 Wood et al 2012) The com-plete list of taxa 363 terminals in total and theGenBank accession numbers are listed in Table S1Taxon names and nomenclatural changes are discussedin the ldquoSystematics of Araneoidea and Nicodamoideardquosection
Molecular methods
For each specimen up to three legs were used fortotal DNA extraction using the DNeasy tissue kit(Qiagen Valencia CA USA) the remainder of thespider was kept as a voucher Purified genomic DNAwas used as a template in order to target the followingsix genes or gene fragments two nuclear ribosomalgenes 18S rRNA (18S hereafter ~1800 bp) and 28SrRNA (28S hereafter fragment of ~2700 bp) twomitochondrial ribosomal genes 12S rRNA (12S here-after ~400 bp) and 16S rRNA (16S hereafter~550 bp) the nuclear protein-encoding gene histoneH3 (H3 hereafter 327 bp) and the mitochondrial pro-tein-encoding gene cytochrome c oxidase subunit I(COI hereafter 771 bp) We did not generate addi-tional wingless sequences as part of the current studyAll wingless sequences used in the analyses come fromprevious studies and were already available in Gen-Bank The PCRs were carried out using IllustraTMpuReTaq Ready-To-Go PCR beads (GE HealthcareUK wwwgelifesciencescom) as described in theSupporting InformationPCR-amplified products were sent to the High
Throughput Sequencing (htSEQ) Genomics Centerfacility at the University of Washington (Seattle WAUSA) for enzymatic cleanup and double-strandedsequencing The resulting chromatograms were readand edited and overlapping sequence fragments assem-bled visually inspected and edited using Sequencherv47 (Gene Codes Corporation Ann Harbor MIUSA) and Geneious v605 (Biomatters available athttpwwwgeneiouscom) In order to detect contam-ination individual fragments were submitted toBLAST (Basic Local Alignment Search Tool) asimplemented on the NCBI website (httpblastncbinlmnihgov) A consensus was compiledfrom all sequenced DNA fragments for each gene andtaxon and deposited in GenBank (Table S1) The bio-logical sequence alignment editor Bioedit v7111(Hall 1999 available at httpwwwmbioncsueduBioEditbioedithtml) was used to edit the completesequences
Phylogenetic analyses
All molecular phylogenetic analyses were run on theAbel Cluster at the University of Oslo the CIPRESscience gateway (Miller et al 2011) and at a Linux
server at the Natural History Museum Oslo Parsi-mony analyses were run on a fast desktop computer atthe Natural History Museum of Denmark Universityof Copenhagen
Alignments Multiple sequence alignments werecarried out with MAFFT v7058b (Katoh andStandley 2013) run on the Ubuntu server at theNatural History Museum University of OsloAlignments of protein-encoding genes were trivial dueto the lack of gaps (except few insertionsdeletions inwingless) and were produced using the L-INS-imethod Ribosomal genes however contain variableregions In addition the distribution of insertions anddeletions is nonrandom in stem regions due tostructural constraints such as compensatory mutationsand consequently taking rRNA secondary structureinto consideration is also important (Rix et al 2008Murienne et al 2010) To that end we have used theQ-INS-i method which implements the four-wayconsistency objective function (Katoh and Toh 2008)Because the Q-INS-i method is computationally verydemanding long fragments such as 18S and 28S werealigned in shorter blocks (based on amplicon limits)which were assembled after alignmentIn a few cases sequences were found to be a con-
tamination or potential paralogues and were excludedfrom the final analyses (see supporting information)However to exemplify the effect of indiscriminatelyincluding all data we ran a round of maximum-likeli-hood (ML) analyses keeping these sequences Theseresults are not discussed further here but are shown inFig S1 Additional data sets were created using differ-ent approaches to improve data completeness ordecrease potential ambiguities To increase data com-pleteness we excluded taxa that were not sequencedfor most of the genes in a stepwise fashion retainingtaxa with data for at least three genes and taxa withdata for at least four genes In order to reduceambiguously aligned regions in the data set we pro-cessed the ribosomal genes with the program trimalv13 (Capella-Gutierrez et al 2009) using the heuris-tic automated1 method and the gappyout method forthe 28S1 fragment for which automated1 failed to pro-vide plausible solution The list of all matrices and thetreatments that were applied to generate them aresummarized in Table S2
Maximum-likelihood The ML analyses were carriedout with the program RAxML (Stamatakis 2014) onCIPRES or on Abel The concatenated gene matrixwas partitioned by gene and the protein-encodinggenes were further partitioned into 1st + 2nd positionand 3rd position partitions Bootstrap and optimaltrees were computed in the same run using the faoption using 1000 bootstrapping replicates Trees were
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 225
rooted using the mygalomorph spider Euagruschisoseus (Dipluridae)
Nonparametric methods and mixture models Becauseeach position in a gene can be under different selectivepressures a site-specific approach to the estimation ofsubstitution rates and other model parameters may bemost appropriate To investigate the effects of thisapproximation we used the nonparametric models ofsite-specific rates of equilibrium frequency profiles asimplemented in PhyloBayes v33e (Lartillot et al2009) We used the CAT-GTR model which is themost appropriate for DNA (-cat -gtr -dgam 4) Twoindependent runs were launched and checked forconvergence and the results are summarized in thetopology presented in Fig S2
Parsimony methods The parsimony analyses of theconcatenated molecular matrix were carried out withthe computer program TNT v11 (Goloboff et al2008) Given the size of the matrix (363 taxa and 7genes) a driven search combining new technologyalgorithms using equal weights (ie tree drifting mixedsectorial searches and tree fusing) was performed (50initial addition sequences initial level 10 cycles ofdrifting 10) until it stabilized onto a strict consensusfive times (with default factor of 75) This is one of themost efficient search strategies when dealing withlarge difficult data sets (Goloboff 1999) Most othersearch settings were left as default values Commandsused were included in and run from a script filewhich was generated by modifying an automaticallygenerated TNT batch file The detailed sequence ofcommands is given in the Supporting InformationNodal support was estimated via 1000 replicates of
parsimony jackknifing (Farris et al 1996 Farris1997) under new technology (using default values)
Divergence time estimation In order to estimatedivergence times we used a relaxed uncorrelatedlognormal approximation (Drummond et al 2006) asimplemented in the program BEAST v211(Bouckaert et al 2014) Analyses in BEAST were runwith exponential distribution for the probabilitydensity of the tmrca prior and birthndashdeath model forthe tree prior Calibration points and relevant priorparameters are listed in Table S3 Parameters werechosen in such a way that 95 of the priorsrsquodistributions fell between the minimum (the offset) andthe maximum values reported for the datinguncertainty of the corresponding fossil Because it isunknown how far the fossil is from the most recentcommon ancestor of the node that it is constraining(eg what is its position along the stem) we used anoninformative hyper prior with gamma distribution toincorporate the uncertainty of the calibration-density
(Heath 2012) All constraints were applied as stemcalibrations In the results presented here we have notincluded as a constraint the fossil spiderMongolarachne jurassica (Selden et al 2011 2013formerly classified as a Nephila species) from theMiddle Jurassic deposits of China (Inner MongoliaDaohugou China) because of recent concerns aboutits taxonomic placement (eg Kuntner et al 2013)However the fossil described by Selden et al (2011)does seem to have morphological characters compatiblewith those of other nephilids A male specimendescribed two years later was assigned to the samespecies (Selden et al 2013) and because the male didnot fit the Nephilidae diagnosis the female (describedas N jurassica) and the male were placed in a newfamilymdashMongolarachnidae Selden et al (2013) didnot present convincing evidence that these twospecimens are conspecific (eg the male resemblesEctatosticta a hypochilid genus endemic to China) soin our view the question of where M jurassica belongsis still in need of further research For example recentdescription of Geratonephila burmanica from EarlyCretaceous Burmese amber (97ndash110 Myr old Poinarand Buckley 2012 see also Penney 2014) challengesthe hypothesis of Kuntner et al (2013) that the cladeof Nephila and its close relatives is only 40ndash60 Myr oldAs a starting tree in all BEAST runs we used the
best tree from the ML analysis of the full data set thatwas processed with the program treePL (Smith andOrsquoMeara 2012) and the same sets of calibration con-straints as for the corresponding BEAST analysesNodes where fossil calibrations were applied were alsoconstrained as monophyletic (note that these werealready selected in order to reflect well-supportedmonophyletic groups as found by the ML analysessee arrows on Fig 3) however the starting tree topol-ogy was not strictly constrained in order to accountfor topological uncertainties Conversion of the MLtree to ultrametric with treePL was necessary in orderto provide BEAST with a starting tree that satisfies allpriors and topological constraints Clock and substitu-tion models were unlinked between gene partitionsexcept for the mitochondrial genes (16S and COI)Analyses were run for at least 200 million generationswith second runs for at least 70 million generations totest for convergence of the results Chain mixing effec-tive sample sizes of estimates and other relevant statis-tics were evaluated in Tracer v15 (Rambaut andDrummond 2007) Trees were summarized with theprogram TreeAnnotator which is distributed as partof the BEAST package Two different sets of datinganalyses were run with calibrations applied in such away that the nephilids are treated as a clade with ara-neids (Araneidae) and as an independent clade (seediscussion in the ldquoSystematics of Araneoidea andNicodamoideardquo section) In addition to the partitioned
226 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Dimitrov et al 2012 Wood et al 2012) The com-plete list of taxa 363 terminals in total and theGenBank accession numbers are listed in Table S1Taxon names and nomenclatural changes are discussedin the ldquoSystematics of Araneoidea and Nicodamoideardquosection
Molecular methods
For each specimen up to three legs were used fortotal DNA extraction using the DNeasy tissue kit(Qiagen Valencia CA USA) the remainder of thespider was kept as a voucher Purified genomic DNAwas used as a template in order to target the followingsix genes or gene fragments two nuclear ribosomalgenes 18S rRNA (18S hereafter ~1800 bp) and 28SrRNA (28S hereafter fragment of ~2700 bp) twomitochondrial ribosomal genes 12S rRNA (12S here-after ~400 bp) and 16S rRNA (16S hereafter~550 bp) the nuclear protein-encoding gene histoneH3 (H3 hereafter 327 bp) and the mitochondrial pro-tein-encoding gene cytochrome c oxidase subunit I(COI hereafter 771 bp) We did not generate addi-tional wingless sequences as part of the current studyAll wingless sequences used in the analyses come fromprevious studies and were already available in Gen-Bank The PCRs were carried out using IllustraTMpuReTaq Ready-To-Go PCR beads (GE HealthcareUK wwwgelifesciencescom) as described in theSupporting InformationPCR-amplified products were sent to the High
Throughput Sequencing (htSEQ) Genomics Centerfacility at the University of Washington (Seattle WAUSA) for enzymatic cleanup and double-strandedsequencing The resulting chromatograms were readand edited and overlapping sequence fragments assem-bled visually inspected and edited using Sequencherv47 (Gene Codes Corporation Ann Harbor MIUSA) and Geneious v605 (Biomatters available athttpwwwgeneiouscom) In order to detect contam-ination individual fragments were submitted toBLAST (Basic Local Alignment Search Tool) asimplemented on the NCBI website (httpblastncbinlmnihgov) A consensus was compiledfrom all sequenced DNA fragments for each gene andtaxon and deposited in GenBank (Table S1) The bio-logical sequence alignment editor Bioedit v7111(Hall 1999 available at httpwwwmbioncsueduBioEditbioedithtml) was used to edit the completesequences
Phylogenetic analyses
All molecular phylogenetic analyses were run on theAbel Cluster at the University of Oslo the CIPRESscience gateway (Miller et al 2011) and at a Linux
server at the Natural History Museum Oslo Parsi-mony analyses were run on a fast desktop computer atthe Natural History Museum of Denmark Universityof Copenhagen
Alignments Multiple sequence alignments werecarried out with MAFFT v7058b (Katoh andStandley 2013) run on the Ubuntu server at theNatural History Museum University of OsloAlignments of protein-encoding genes were trivial dueto the lack of gaps (except few insertionsdeletions inwingless) and were produced using the L-INS-imethod Ribosomal genes however contain variableregions In addition the distribution of insertions anddeletions is nonrandom in stem regions due tostructural constraints such as compensatory mutationsand consequently taking rRNA secondary structureinto consideration is also important (Rix et al 2008Murienne et al 2010) To that end we have used theQ-INS-i method which implements the four-wayconsistency objective function (Katoh and Toh 2008)Because the Q-INS-i method is computationally verydemanding long fragments such as 18S and 28S werealigned in shorter blocks (based on amplicon limits)which were assembled after alignmentIn a few cases sequences were found to be a con-
tamination or potential paralogues and were excludedfrom the final analyses (see supporting information)However to exemplify the effect of indiscriminatelyincluding all data we ran a round of maximum-likeli-hood (ML) analyses keeping these sequences Theseresults are not discussed further here but are shown inFig S1 Additional data sets were created using differ-ent approaches to improve data completeness ordecrease potential ambiguities To increase data com-pleteness we excluded taxa that were not sequencedfor most of the genes in a stepwise fashion retainingtaxa with data for at least three genes and taxa withdata for at least four genes In order to reduceambiguously aligned regions in the data set we pro-cessed the ribosomal genes with the program trimalv13 (Capella-Gutierrez et al 2009) using the heuris-tic automated1 method and the gappyout method forthe 28S1 fragment for which automated1 failed to pro-vide plausible solution The list of all matrices and thetreatments that were applied to generate them aresummarized in Table S2
Maximum-likelihood The ML analyses were carriedout with the program RAxML (Stamatakis 2014) onCIPRES or on Abel The concatenated gene matrixwas partitioned by gene and the protein-encodinggenes were further partitioned into 1st + 2nd positionand 3rd position partitions Bootstrap and optimaltrees were computed in the same run using the faoption using 1000 bootstrapping replicates Trees were
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 225
rooted using the mygalomorph spider Euagruschisoseus (Dipluridae)
Nonparametric methods and mixture models Becauseeach position in a gene can be under different selectivepressures a site-specific approach to the estimation ofsubstitution rates and other model parameters may bemost appropriate To investigate the effects of thisapproximation we used the nonparametric models ofsite-specific rates of equilibrium frequency profiles asimplemented in PhyloBayes v33e (Lartillot et al2009) We used the CAT-GTR model which is themost appropriate for DNA (-cat -gtr -dgam 4) Twoindependent runs were launched and checked forconvergence and the results are summarized in thetopology presented in Fig S2
Parsimony methods The parsimony analyses of theconcatenated molecular matrix were carried out withthe computer program TNT v11 (Goloboff et al2008) Given the size of the matrix (363 taxa and 7genes) a driven search combining new technologyalgorithms using equal weights (ie tree drifting mixedsectorial searches and tree fusing) was performed (50initial addition sequences initial level 10 cycles ofdrifting 10) until it stabilized onto a strict consensusfive times (with default factor of 75) This is one of themost efficient search strategies when dealing withlarge difficult data sets (Goloboff 1999) Most othersearch settings were left as default values Commandsused were included in and run from a script filewhich was generated by modifying an automaticallygenerated TNT batch file The detailed sequence ofcommands is given in the Supporting InformationNodal support was estimated via 1000 replicates of
parsimony jackknifing (Farris et al 1996 Farris1997) under new technology (using default values)
Divergence time estimation In order to estimatedivergence times we used a relaxed uncorrelatedlognormal approximation (Drummond et al 2006) asimplemented in the program BEAST v211(Bouckaert et al 2014) Analyses in BEAST were runwith exponential distribution for the probabilitydensity of the tmrca prior and birthndashdeath model forthe tree prior Calibration points and relevant priorparameters are listed in Table S3 Parameters werechosen in such a way that 95 of the priorsrsquodistributions fell between the minimum (the offset) andthe maximum values reported for the datinguncertainty of the corresponding fossil Because it isunknown how far the fossil is from the most recentcommon ancestor of the node that it is constraining(eg what is its position along the stem) we used anoninformative hyper prior with gamma distribution toincorporate the uncertainty of the calibration-density
(Heath 2012) All constraints were applied as stemcalibrations In the results presented here we have notincluded as a constraint the fossil spiderMongolarachne jurassica (Selden et al 2011 2013formerly classified as a Nephila species) from theMiddle Jurassic deposits of China (Inner MongoliaDaohugou China) because of recent concerns aboutits taxonomic placement (eg Kuntner et al 2013)However the fossil described by Selden et al (2011)does seem to have morphological characters compatiblewith those of other nephilids A male specimendescribed two years later was assigned to the samespecies (Selden et al 2013) and because the male didnot fit the Nephilidae diagnosis the female (describedas N jurassica) and the male were placed in a newfamilymdashMongolarachnidae Selden et al (2013) didnot present convincing evidence that these twospecimens are conspecific (eg the male resemblesEctatosticta a hypochilid genus endemic to China) soin our view the question of where M jurassica belongsis still in need of further research For example recentdescription of Geratonephila burmanica from EarlyCretaceous Burmese amber (97ndash110 Myr old Poinarand Buckley 2012 see also Penney 2014) challengesthe hypothesis of Kuntner et al (2013) that the cladeof Nephila and its close relatives is only 40ndash60 Myr oldAs a starting tree in all BEAST runs we used the
best tree from the ML analysis of the full data set thatwas processed with the program treePL (Smith andOrsquoMeara 2012) and the same sets of calibration con-straints as for the corresponding BEAST analysesNodes where fossil calibrations were applied were alsoconstrained as monophyletic (note that these werealready selected in order to reflect well-supportedmonophyletic groups as found by the ML analysessee arrows on Fig 3) however the starting tree topol-ogy was not strictly constrained in order to accountfor topological uncertainties Conversion of the MLtree to ultrametric with treePL was necessary in orderto provide BEAST with a starting tree that satisfies allpriors and topological constraints Clock and substitu-tion models were unlinked between gene partitionsexcept for the mitochondrial genes (16S and COI)Analyses were run for at least 200 million generationswith second runs for at least 70 million generations totest for convergence of the results Chain mixing effec-tive sample sizes of estimates and other relevant statis-tics were evaluated in Tracer v15 (Rambaut andDrummond 2007) Trees were summarized with theprogram TreeAnnotator which is distributed as partof the BEAST package Two different sets of datinganalyses were run with calibrations applied in such away that the nephilids are treated as a clade with ara-neids (Araneidae) and as an independent clade (seediscussion in the ldquoSystematics of Araneoidea andNicodamoideardquo section) In addition to the partitioned
226 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
rooted using the mygalomorph spider Euagruschisoseus (Dipluridae)
Nonparametric methods and mixture models Becauseeach position in a gene can be under different selectivepressures a site-specific approach to the estimation ofsubstitution rates and other model parameters may bemost appropriate To investigate the effects of thisapproximation we used the nonparametric models ofsite-specific rates of equilibrium frequency profiles asimplemented in PhyloBayes v33e (Lartillot et al2009) We used the CAT-GTR model which is themost appropriate for DNA (-cat -gtr -dgam 4) Twoindependent runs were launched and checked forconvergence and the results are summarized in thetopology presented in Fig S2
Parsimony methods The parsimony analyses of theconcatenated molecular matrix were carried out withthe computer program TNT v11 (Goloboff et al2008) Given the size of the matrix (363 taxa and 7genes) a driven search combining new technologyalgorithms using equal weights (ie tree drifting mixedsectorial searches and tree fusing) was performed (50initial addition sequences initial level 10 cycles ofdrifting 10) until it stabilized onto a strict consensusfive times (with default factor of 75) This is one of themost efficient search strategies when dealing withlarge difficult data sets (Goloboff 1999) Most othersearch settings were left as default values Commandsused were included in and run from a script filewhich was generated by modifying an automaticallygenerated TNT batch file The detailed sequence ofcommands is given in the Supporting InformationNodal support was estimated via 1000 replicates of
parsimony jackknifing (Farris et al 1996 Farris1997) under new technology (using default values)
Divergence time estimation In order to estimatedivergence times we used a relaxed uncorrelatedlognormal approximation (Drummond et al 2006) asimplemented in the program BEAST v211(Bouckaert et al 2014) Analyses in BEAST were runwith exponential distribution for the probabilitydensity of the tmrca prior and birthndashdeath model forthe tree prior Calibration points and relevant priorparameters are listed in Table S3 Parameters werechosen in such a way that 95 of the priorsrsquodistributions fell between the minimum (the offset) andthe maximum values reported for the datinguncertainty of the corresponding fossil Because it isunknown how far the fossil is from the most recentcommon ancestor of the node that it is constraining(eg what is its position along the stem) we used anoninformative hyper prior with gamma distribution toincorporate the uncertainty of the calibration-density
(Heath 2012) All constraints were applied as stemcalibrations In the results presented here we have notincluded as a constraint the fossil spiderMongolarachne jurassica (Selden et al 2011 2013formerly classified as a Nephila species) from theMiddle Jurassic deposits of China (Inner MongoliaDaohugou China) because of recent concerns aboutits taxonomic placement (eg Kuntner et al 2013)However the fossil described by Selden et al (2011)does seem to have morphological characters compatiblewith those of other nephilids A male specimendescribed two years later was assigned to the samespecies (Selden et al 2013) and because the male didnot fit the Nephilidae diagnosis the female (describedas N jurassica) and the male were placed in a newfamilymdashMongolarachnidae Selden et al (2013) didnot present convincing evidence that these twospecimens are conspecific (eg the male resemblesEctatosticta a hypochilid genus endemic to China) soin our view the question of where M jurassica belongsis still in need of further research For example recentdescription of Geratonephila burmanica from EarlyCretaceous Burmese amber (97ndash110 Myr old Poinarand Buckley 2012 see also Penney 2014) challengesthe hypothesis of Kuntner et al (2013) that the cladeof Nephila and its close relatives is only 40ndash60 Myr oldAs a starting tree in all BEAST runs we used the
best tree from the ML analysis of the full data set thatwas processed with the program treePL (Smith andOrsquoMeara 2012) and the same sets of calibration con-straints as for the corresponding BEAST analysesNodes where fossil calibrations were applied were alsoconstrained as monophyletic (note that these werealready selected in order to reflect well-supportedmonophyletic groups as found by the ML analysessee arrows on Fig 3) however the starting tree topol-ogy was not strictly constrained in order to accountfor topological uncertainties Conversion of the MLtree to ultrametric with treePL was necessary in orderto provide BEAST with a starting tree that satisfies allpriors and topological constraints Clock and substitu-tion models were unlinked between gene partitionsexcept for the mitochondrial genes (16S and COI)Analyses were run for at least 200 million generationswith second runs for at least 70 million generations totest for convergence of the results Chain mixing effec-tive sample sizes of estimates and other relevant statis-tics were evaluated in Tracer v15 (Rambaut andDrummond 2007) Trees were summarized with theprogram TreeAnnotator which is distributed as partof the BEAST package Two different sets of datinganalyses were run with calibrations applied in such away that the nephilids are treated as a clade with ara-neids (Araneidae) and as an independent clade (seediscussion in the ldquoSystematics of Araneoidea andNicodamoideardquo section) In addition to the partitioned
226 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses we also ran an analysis treating the wholedata set as a single partition This was done in orderto compare both approaches and because it has beenshown that in some cases partitioning may cause sta-tistical problems in dating analyses (eg Dos Reiset al 2014)
Comparative analyses
We used the web architecture data matrix fromDimitrov et al (2012) as a base for the current analy-ses Additional taxa were added to this data set anddespite the number of species with unknown webarchitecture representatives from all orb-weaving fam-ilies were scored in the data set (the web charactermatrix is available as supporting information) Com-parative analyses were carried out using the ultramet-ric trees from the dating analyses and the R packagesape (Paradis 2012) and phytools (Revell 2012) Likeli-hood models for discrete characters may be based onthree general assumptions about the rates of charactertransformation (1) equal rates of transition betweenstates (ER) (2) a symmetric model where forward andreverse rates of transition between two states are equalbut other rates may vary (SYM) and (3) the mostparameterized case of all rates being different (ARD)We fitted these three models to our data and selectedthe one that resulted in the highest likelihood To dothis we used the function ace in ape with type = ldquodis-creterdquo The best-performing model was then used toreconstruct web evolution using a stochastic charactermapping approach (SIMMAP) as implemented in phy-tools (with the makesimmap function) A thousandstochastic maps were generated using 1000 values forthe Q matrix obtained from the posterior distributionusing the Q = ldquomcmcrdquo command and nsim = 1000 asa prior and results were summarized on the corre-sponding BEAST summary tree The stochastic char-acter mapping is a Bayesian approximation toancestral state reconstruction (Bollback 2006) Wepreferred SIMMAP to other likelihood approaches toancestral state reconstruction of discrete traits becauseit allows changes to occur along branches and forassessing the uncertainty in character historyIn addition to web architecture we also scored the
presence or absence of a cribellum for all taxa in ourmatrix The cribellum is a part of a complex spinningapparatus present in all cribellate spiders regardless oftheir web architecture For example some cribellatesbuild orb-webs whereas others may build sheet orirregular webs The presence of the calamistrum (afourth metatarsus comb made out of modifiedmacrosetae) as well as a diversity of silk ldquocombingrdquobehaviours are correlated with the cribellum in theproduction of the cribellate silk that we observe intheir webs In earlier classification systems the
presence or absence of a cribellum had been used asan important diagnostic character separating araneo-morph spiders into two large groupsmdashcribellates andecribellates This early view has been replaced by thecurrent paradigm of cribellum evolution which treatsthis character system (and the associated cribellateweb) as a symplesiomorphic araneomorph feature thathas undergone multiple losses during the evolutionaryhistory of this lineage (eg Lehtinen 1967 Griswoldet al 1999 2005 Spagna and Gillespie 2008 Milleret al 2010) The most recent study of cribellum evolu-tion (Miller et al 2010) used a large sample of arane-omorph lineages and parsimony and Bayesianmethods to infer the history of this character Becauseof the complexity of the cribellate spinning apparatusMiller et al (2010) argued that it is likely to expectthat rates of transition between character states areasymmetrical for these particular characters Althoughthis is a plausible expectation in their analyses theyhad to manually alter rates of character transforma-tion in order to find a minimum threshold at whichthe cribellum is reconstructed as symplesiomorphic inaraneomorphs that is with a single origin and theimplied multiple losses They also suggested that addi-tional data might improve the results reconstructingthe cribellum as homologous and allowing for actualestimation of the rates of cribellum gain and loss Weagree with the arguments for rates asymmetry pre-sented in Miller et al (2010) and here we test if thecombined use of a different approach to ancestral statereconstruction with a larger data set is capable of fur-ther elucidating this problem The methods used tostudy the evolution of the cribellum are the same asthose described above for web architecture
Results
The ML analyses of the full data set (Figs 2 S3)recover Araneoidea as a clade with Nicodamoidea asits sister group both with a bootstrap support gt 75(bootstrap support values are given in Table S4 andalso shown on Figs 2 S3) The monophyly of cribel-late and ecribellate nicodamids receives high supportand this clade is what we now rank as the superfamilyNicodamoideaThe clade that includes both the cribellate and
ecribellate orb-weavers also includes the RTA cladeOecobiidae and Hersiliidae and is the sister group to amonophyletic Eresidae albeit with low support Thesuperfamily Deinopoidea is paraphyletic with respectto a lineage that includes the RTA clade Hersiliidaeand Oecobiidae Consequently the Orbiculariae arenot monophyletic The cribellate orb-weaving familyUloboridae is monophyletic and well supported and issister group albeit with low support to a lineage that
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 227
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
includes the RTA clade Hersiliidae and OecobiidaeThe monophyly of the RTA clade is well supportedhowever Although lacking nodal support in the opti-mal tree Deinopidae is sister group to a lineage thatincludes Uloboridae (Hersiliidae + Oecobiidae) andthe RTA clade Deinopidae is well supported
The results show high support for the monophyly ofmost Araneoidea families with a few exceptions Ingeneral bootstrap support values improve when parti-tion completeness is optimized (see Table S4 and FigsS4 S5) Anapidae includes Anapis the micropholcom-matines and the holarchaeids the family is never
Fig 2 Summary of topologies and clade supports from the different phylogenetic analyses described in the materials and methods sectionFamily crown groups are collapsed into coloured triangles Most triangles are equally sized their sizes are not proportional to the number ofrepresentatives included in the analyses (a total of 363 terminals were included in the analyses) The base topology is the maximum-likelihood(ML) result from the analyses of the complete data set Black squares denote ML bootstrap values gt70 grey squares indicate maximum parsi-mony (MP) bootstrap value gt 70 and black stars show posterior probabilities from the PhyloBayes analyses which are ge 95 Alternativetopologies are shown on the right black arrows correspond to PhyloBayes results and blue arrows show alternative ML resolutions Because theMP tree showed more differences these are not summarized here but the full MP topology is available in Fig S7
228 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
recovered as monophyletic even if Holarchaea is con-sidered an anapid because a second ldquoanapidrdquo cladecomprising Gertschanapis Maxanapis and Chasmo-cephalon resolves elsewhere The family Synotaxidaeappears as diphyletic because the synotaxines are notclosely related to the pahorine + physoglenine cladeHowever the monophyly of the latter two subfamiliesas a clade is well supportedLinyphiidae plus Pimoidae form a clade but neither
family is supported as monophyletic due to the cluster-ing of the Asian pimoid genera Weintrauboa andPutaoa with the early branching linyphiid genus Ste-monyphantes (this clade is strongly supported) Sup-port values for most nodes at the base of linyphioids(Linyphiidae plus Pimoidae) are low as well as that ofthe node that indicates that the sister group of lsquoliny-phioidsrsquo is the Physogleninae plus Pahorinae synotaxidclade (which we group now under the family namePhysoglenidae)Nodal support for interfamilial relationships is gener-
ally low across Araneoidea except in a few instancesthe clade of Mimetidae plus Arkyidae + Tetragnathi-dae and the clade of Malkaridae plus PararchaeidaeThe arkyines (which we rank at the family level in ourrevised classification) represented here by nine termi-nals are monophyletic and well supported but do notfall within Araneidae (where they are currently classi-fied) instead the arkyine clade is sister group to Tetrag-nathidae and this lineage is sister to MimetidaeNephilidae plus Araneidae form a well-supported cladeand although both groups appear reciprocally mono-phyletic in some analyses nodal support for Araneidaeis low whereas it is high for the clade of Nephila and itsclosest relatives The symphytognathoid families consti-tute a polyphyletic group although all the nodesinvolving these interfamilial relationships receive lowsupport values Cepheia longiseta the single representa-tive of Synaphridae in our analyses is sister group tothe Symphytognathidae lineageThe ML analyses of the data sets where ambigu-
ously aligned blocks of data were excluded (matrix_tri-mal) and those based on data sets where taxa with lowgene representation were excluded (matrix_3g and ma-trix_4g) recovered results that were highly congruentwith those from the full data set Different resolutionsinvolved only groupings that received lower supportand did not involve any of the clades discussed aboveResults from these analyses are summarized in Fig 2and full topologies are presented in Figs S4ndashS6 Giventhis high congruence of the results from different datatreatments we used only the full data set (as it con-tains the highest amount of data and retains all taxa)for the Bayesian and parsimony analysesResults from PhyloBayes (Fig S2) are highly congru-
ent with those from ML except for a handful ofinstances that are highlighted on Fig 2 From those
the most significant are the recovery of a monophyleticAnapidae that includes Holarchaeidae and the move ofCyatholipidae to a clade together with PimoidaeLinyphiidae and Synaphridae Parsimony analyses inTNT found 211 shortest trees and after collapsing andfiltering out zero length branches a single tree wasretained (shown in Fig S7) TNT results are mostlycongruent with ML and Bayesian results but the sup-port for some groups is lower showing once more thatthe amount of information available to resolve thesefamilies is limited particularly at the interfamilial anddeeper levels Only some of the interfamilial groupingssuch as the clade [Mimetidae + (Arkyidae + Tetrag-nathidae)] were recovered with high support
Molecular dating results
The annotated highest clade credibility tree from theBEAST analyses with dating scheme applying the oldestfossil described as araneid to Araneidae sl is presentedin Fig 3 Additional trees from the different BEASTruns are available as supporting information (Figs S8and S9) The results showed convergence for most of theparameters but in some cases effective sampling sizes(ESS) of relevant estimates were not optimal (higherthan 150 but less than 200) Independent runs of datinganalyses showed a tendency to converge but because ofthe size of the current data set and the time required torun a large number of generations only one instance ofeach analysis was allowed to sample more than 200 mil-lion states from the posterior distribution Close exami-nations of the results and lack of improvement whenextending the sampling suggest that many of these prob-lems are likely due to topological uncertainties in combi-nation with missing data The best example for this isthe case of Pimoa and the clade Pimoa + Nanoa inwhich the estimate for the age of its stem varies signifi-cantly between the two most common topologies pre-sented in the posterior sample either as sister group tothe other pimoids + linyphiids or as closely related tophysoglenids As expected different dating strategiesand use of partitioned versus unpartitioned analysesresulted in slightly different age estimatesDespite these differences in the inferred median ages
95 intervals of probability densities from all analysesare congruent and show overlap It is worthwhilespecifically mentioning the case of nephilids becausethey have been the subject of a detailed study recently(Kuntner et al 2013) In our analyses we did notimplement a constraint for this group due to theunclear status of some of the available fossils The ageof Nephila in all of our analyses was found to beyounger than that suggested by Mongolarachne juras-sica and the estimated age of the genus and the wholesubfamily was closer to the estimates of Kuntner et al(2013) The median ages from our unpartitioned
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 229
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Philoponella variabilis
Wadotes dixiensis
Acrobleps sp 002AUST
Argyroneta aquatica
Dresserus kannemeyeri
Callobius sp
Epeirotypus brevipes
Novalena intermedia
Oecobius sp
Thwaitesia sp
Dictyna sp
Argyrodes argentatus
Euryopis funebris
Ambicodamus marae
Phycosoma mustelinum
Neoscona arabesca
Gnolus sp GH1020
Argiope trifasciata
Desis formidabilis
Zorocrates fuscus
Megadictyna thilenii
Alopecosa kochi
Caerostris sp 1248
Amaurobius similis
Ariamnes attenuata
Mallos pallidus
Helvibis cf longicauda
Hyptiotes gertschi
Stegodyphus lineatus
Metaltella simoni
Deliochus sp
Platnickia alabamensis
Echinotheridion otlum
Styposis selis
Mecynogea lemniscata
Oncodamus bidensAmbicodamus sp
Stegodyphus mimosarum
Anelosimus nigrescens
Deinopis sp
Caerostris sp 1230
Gnolus sp GH1023
Uroctea durandi
Zodarion sp
Taphiassa sp Qsld Rix
Paraphidippus aurantius
Coelotes terrestris COET13
Corinnidae PAN
Mexitlia trivittata
Ambohima sp
Nesticodes rufipes
Dolomedes tenebrosus
Waitkera waitakerensis
Taira sp
Zygiella x notata
Stiphidion facetum
Gandanameno fumosa
Synotaxus waiwai
Calymmaria sp
Neolana dalmasi
Coleosoma acutiventer
Araneus diadematus
Synotaxus sp 1385
Tengella radiata
Theridiosomatidae NN
Clitaetra sp
Caerostris sp 1243
Hersiola macullata
Chrysso albipes
Steatoda bipunctata
Cybaeolus sp
Eresus walckenaeri
Cybaeus morosus
Nicodaminae
Dorceus fastuosus
Seothyra annettae
Agelenopsis aperta
Anelosimus baeza
Achaearanea tepidariorum
Neottiura bimaculata
Oncodamus decipiens
Metepeira labyrinthea
Spintharus flavidus
Dipoena cf hortoni
Anelosimus analyticus
Eresus sp nov
Taphiassa sp
Nephilengys malabarensis
Episinus angulatus
Penestomus sp
Latrodectus geometricus
Cyclosa conica
Neoramia janus
Acanthepeira stellata
Clitaetra perroti
Zelotes sp
Eurocoelotes inermis
Cryphoeca sp
Gasteracantha cancriformis
Nephila clavipes
Mastophora phrynosoma
Steatoda borealis
Anapisona kethleyi
Anelosimus domingo
Holarchaea sp
Oarces reticulatus 1014
Textrix denticulata
Hersilia insulana
Theridion acutitarse
Zosis sp
Argiope argentata
Argyrodes trigonum
Uloborus diversus
Peucetia viridans
Cavernocymbium prentoglei
Dresserus colsoni
Stegodyphus annulipes
cf Aschema sp
Phoroncidia americana
Taphiassa punctata
Stegodyphus tentoriicola
Taphiassa sp Tasm Rix
Cyrtophora moluccensis
Zygiella atrica
Barronopsis barrowsi
Anyphaena californica
Deinopis spinosa
Tidarren sisyphoides
Oarces sp
Tegenaria domestica
Vidole capensis
Menneus sp
Enoplognatha caricis
Tamgrinia alveolifera
Agelena gracilens
Zanomys californica
Gandanameno spenceri
Micrathena gracilis
Chrosiothes cf jocosus
Larinioides cornutus
Pimus sp
Thymoites unimaculatus
Chumma inquieta
Rhomphaea metalissima
Phonognatha graeffei
Mangora maculata
Eresus cf kollari
Araneus marmoreus
Herennia multipuncta
Stegodyphus sp
Theridion varians
Adonea fimbriata
Yunohamella lyricus
Holarchaea sp ARACG000249
Robertus neglectus
Acrobleps hygrophilus
Chresiona sp
Pholcomma hirsutum
Hahnia clathrata
Argiope savignyi
Uloborus glomosus
Anapis sp 1206
Badumna longiqua
Gandanameno sp
Cerocida strigosa
Deinopis sp 1160
QNeogenePaleogeneCretaceousJurassicTriassic
02623661452013Forstera sp
Malkara sp GH1221
Hispanognatha guttata
Meioneta rurestris
Perissopmeros sp 1587
Symphytognathidae 005AUST
Pachygnatha degeeri
Helophora insignis
Eryciniolia purpurapunctata
Cyatholipidae
Mollemeta edwardsi
Dolichognatha sp
Mysmeninae 033 MAD
Dubiaranea aysenensis
Mangua gunni
Azilia guatemalensis
Pahora mrijiku
Azilia sp GH0834
Arkys sp 1102
Malkaridae sp GH1720
Chilenodes sp 1229
Tylorida striata
Malkaridae sp GH1207
Malkaridae sp GH1116
Carathea sp GH1093
Matilda sp
Floronia bucculenta
Metabus ebanoverde
Arkys sp 1252
Physoglenes
Mysmena sp GUYANA
Malkara sp GH1220
Mecynometa sp GH0850
Pimoa breuili
Malkara sp GH1158
Australomimetus sp 1115
Linyphia triangularis
Ozarchaea platnicki
Nanometa sp 1137
Meringa sp Otago
Pimoa trifurcata
Alaranea merina
Meta menardi
Chrysometa alajuela
Tupua sp
Diplostyla concolor
Neriene variabilis
Tylorida sp
Mimetus sp 881
Carathea sp
Stemonyphantes abatensis
Ostearius melanopygius
Metellina segmentata
Perissopmeros sp 1588
Nesticella sp 1210
Metainae sp
Oedothorax apicatus
Mimetus sp
Mimetus sp 891
Trogloneta sp 025CHILE
Pararchaea sp
Malkara sp GH1589
Mysmena sp 036THAI
Labulla thoracica
Pimoa sp
Malkara sp N
Tekelloides australis
Tetragnathidae new genus
Archemorus sp 1250
Malkara sp GH998
Agyneta ramosa
Edmanella sp 1599
Arkys cornutus
Maymena ambita
Nanometa sp 1139
Leucauge argyra
Chilenodes australis
Microdipoena nyungwe
Pseudafroneta incerta
Nesticus cellulanus
Malkara sp GH1154
Maxanapis bartle
Lepthyphantes minutus
Malkara sp GH1249
Metainae sp 1
Microlinyphia dana
Archemorus sp 1242
Ero sp 1092
Malkara sp
Edmanella sp 1701
Malkara sp GH1162
Gelanor sp 1605
Meringa borealis
Drapetisca socialis
Tylorida ventralis
Stemonyphantes lineatus
Gertschanapis shantzi
Diphya spinifera
Nesticella sp 1202
Trogloneta sp 024 CHILE
Tekella absidata
Cyrtognatha espaniola
Haplinis diloris
Runga nina
Gongylidiellum vivium
Malkara sp GH1247
Orsinome sp
Allende nigrohumeralis
Australolinyphia remota
Nanoa enana
Pocobletus sp 1387
Chasmocephalon sp
Wanzia sp
Novafroneta vulgaris
Pinkfloydia harveii
Frontinella communis
Cepheia sp
Chileotaxus sp
Weintrauboa chikunii
Metellina merianae
Meta sp 1404
Trogloneta sp 022 ARG
Tenuiphantes tenuis
Perissopmeros sp
Synotaxidae sp
Mysmena sp 037THAI
Nanometa sp 114
Arkys sp 1107
Maymena sp 004MEX
Archemorus sp 1245
Glenognatha sp GH0759
Neriene sp
Patu sp
Pocobletus sp
Orsinome cf vethi
Meta rufolineataPinkfloydia sp
Antillognatha lucida
Teemenaarus sp 1149
Tetragnatha versicolor
Mysmeninae 032 MAD
Notholepthyphantes australis
Meta ovalis
Nanometa sp
Bolyphantes alticeps
Pocobletus sp N
Arkys lancearius
Chilenodes sp 1005
Malkara sp GH1122
Opadometa sp
Archemorus sp 1586
Laetesia raveniLaetesia sp
Mysmeninae 031 MAD
Mesida sp GH0535Tetragnathidae sp
Pimoa
Microneta viaria
Bathyphantes gracilis
Orsonwelles polites
Palaeohyphantes
Orsonwelles malus
Metainae sp 2
Metleucauge sp GH0897
Perissopmeros sp 1109
Putaoa sp 1391
Mysmena sp 013THAI
Microdipoena guttata
Mughiphantes sp 1714
Tetragnatha mandibulata
Gonatium rubellum
Leucauge venusta
Neriene radiata
Symphytognathidae 003MAD
Malkara loricata
Erigone dentosa
QNeogenePaleogeneCretaceousJurassic
0262366145
Fig 3 Results from molecular dating in BEAST using the Araneidae constraint to the redefined Araneidae (including Nephilinae) Grey bars atnodes represent the 95 credibility interval for node age estimates Some outgroup clades that are not discussed in the text are not shown dueto space constraints Black arrows show the branches to which dating constraints were applied (grey arrow shows the branch of the alternativeapplication of the Araneidae constraint see also Fig S8)
230 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
analyses are particularly close to the findings of Kunt-ner et al (2013) Clearly all ldquonephilidrdquo fossils deservefurther study Additional results based on the treefrom the alternative dating scheme for Araneidae arepresented in Fig S10
Web architecture and cribellum evolution
The Araneidae calibration was applied both includ-ing the nephilids and excluding them because thesetwo alternatives result in some slight topological differ-ences and minor discrepancies of the branch lengthestimates of the ultrametric trees For this reason weran comparative analyses on both dated trees Fittingthe three general models for rates of character trans-formation applicable to discrete characters (ER SYMand ARD) on the web architecture data set alwaysresulted in ER giving the highest log-likelihoodBecause conceptually ER is also the simplest modelwe selected these results and ran SIMMAP using theER model SIMMAP results from both topologieswere highly congruent and here we present only theresult from running the analyses with the tree that wasdated with an araneid circumscription that includesthe nephilids (Fig 4)The comparison between ER SYM and ARD models
for the cribellate data resulted in the ARD reconstruc-tion having a slightly better likelihood (although notstatistically significant under the likelihood ratio testmdashv2 P-value of 07148122) Because Miller et al (2010)have discussed at length the arguments for adopting anapproach where the rate of cribellum state transforma-tions are asymmetrical we follow this approach in ourSIMMAP analyses and do not try to further optimizeand achieve higher significance for the ARD results (seeMiller et al 2010 for such results and discussion)Ancestral state reconstruction of the cribellum (andhence the ecribellate web) under an ARD model corrob-orates the homology of this structure and the cribellateweb without ad hoc manipulation of the rates or othermodel parameters The results from the SIMMAP anal-yses using the araneid calibration (including nephilines)are summarized in Fig 5 Additional results based ondated tree using the alternative dating scheme forAraneidae are presented in Fig S11 It is worth men-tioning here that as in previous analyses using ER (seediscussion in Miller et al 2010) our results under ERand SYM models (which are equivalent for a two statecharacter) also contradicted the single origin of thecribellum and the cribellate web
Discussion
In general the phylogenetic signal provided by theanalysed sequences finds support for the monophyly of
most araneoid families as well as for relationshipswithin families Most interfamilial nodes howeverinvolve short internal branches with low nodal sup-port Although some of the relationships with low sup-port values were deemed suspicious in previousSanger-based sequence analyses (such as the placementof the RTA clade among orbicularians) some are nowbeing corroborated by larger transcriptomic analyses(Bond et al 2014 Fernandez et al 2014) This phe-nomenon corroboration of ldquounsupportedrdquo nodesthrough phylogenomics should council against hastilydiscarding topologies simply because of poor supportvaluesIncreased taxon sampling (relative to the taxa used
in Dimitrov et al (2012) the direct predecessor ofthis study) has improved the support values for themonophyly of a few araneoid families (eg Tetrag-nathidae) resolved some controversial placements(eg increased sample of cyatholipids from two toeight representatives has moved out this lineage froman earlier placement within a Linyphiidae + Pimoidaeclade) and supported the circumscription of a fewnew families (eg Arkyidae Physoglenidae) but forthe most part has not resolved araneoid interfamilialrelationships The dating analyses done so far (egAyoub et al 2007 Dimitrov et al 2012 Bond et al2014 this paper) agree in suggesting that the cladoge-netic events and the diversification of araneoid fami-lies are both ancient and compressed in a relativelynarrow time interval (Fig 2) Because most araneoidfamilies were already present during the Cretaceous(Fig 3) we can hypothesize that web architecturessimilar to those that characterize their extant specieswere already diverse at the time of the spectaculardiversification of holometabolous insects (primarilyHymenoptera Diptera and Lepidoptera) (eg Misofet al 2014) which coincide with the angiosperm radi-ation Although in the present study we are notexplicitly testing hypotheses of insectndashspider codiversi-fication (eg Penney 2003) we should point out thatthe findings reported here are concordant with ourprevious hypothesis (Dimitrov et al 2012) suggestingthat the diversification of araneoid webs whichincludes numerous shifts in web architecture and ofweb-building behaviours likely have been driven byenvironmental factors (such as increasing complexityof habitats) availability of prey and intraguild com-petition The subject of orb-weaversrsquo diversificationrequires special attention and we will address it in aseparate paperOur data refute the long-held paradigm of orbicular-
ian monophyly (eg Coddington 1986 Dimitrovet al 2012) by including the RTA clade in the samelineage that groups the cribellate (Deinopoidea) andecribellate (Araneoidea) orb-weavers This latter resultbased on DNA sequence data is by no means new
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 231
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Araneoidea
RTA clade
Uloboridae
Deinopidae
Orb
Brush sheet
Terminal line
Irregular aerial sheet
Irregular ground sheet
Stereotyped aerial sheet
Cobweb
Bolas
No foraging web
Single or few lines in tension
Fig 4 Web architecture evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on the redefined Aranei-dae (including Nephilinae) dating Colours represent different web types sectors of pies at nodes are proportional to the probabilities of eachstate at that node scale is in Myr
232 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
300 250 200 150 100 50 0
Cribellum presentCribellum absent
Fig 5 Cribellum evolutionary history summary of 1000 SIMMAP characters maps using the dated tree based on redefined Araneidae (includ-ing Nephilinae) dating Presence or absence of cribellum is represented by different colours sectors of pies at nodes are proportional to the prob-abilities of each state at that node scale is in Myr
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 233
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
(eg Hayashi 1996 Hausdorf 1999) but has been dis-missed repeatedly in favour of the orbicularian mono-phyly hypothesis (eg Blackledge et al 2009Agnarsson et al 2013) Our results based on the lar-gest sample of orbicularians analysed to date corrobo-rate recent findings about the origin of Orbiculariaewhich used transcriptomic data for a more modesttaxon sample (Bond et al 2014 Fernandez et al2014) Furthermore the results presented herein sug-gest that nicodamids are the closest relatives to a cladethat includes all ecribellate orb-weavers as suggestedin the combined analysis of Blackledge et al (2009)and Dimitrov et al (2012) (see also systematic discus-sion below)
Web architecture and web type evolution
Despite the diversity of web architectures repre-sented by the taxon sample analysed herein (eg seeFigs 1AndashC 6ndash10) the lack of robust nodal support atthe interfamilial level does not allow us to address webarchitecture evolution within Araneoidea satisfactorilyAdditional difficulties stem from the lack of a goodfossil record and uncertainties in the dating and thesystematic circumscription of some of the oldestknown orb-weaver fossils There are however severalgeneral trends that emerge from the results presentedhere The orb-web is ancient having evolved at leastby the early Jurassic By the late Jurassic the orb-web
(A)
(C) (D) (E)
(H)
(F) (G)
(B)
Fig 6 (A) The horizontal sheet-web of an undescribed Cyatholipidae from Australia (DSC_3145) (B) The micro-orb of Tasmanapis strahan(Anapidae) from Tasmania (DSC_0497) (C) The ldquochicken-wirerdquo modular web of Synotaxus sp (Synotaxidae) from Brazil (DSC_9305) (D) Thebowl-shaped sheet-web of an undescribed linyphiid from Taiwan (DSC_0971) (E) Detail of (A) the spider extremely small relative to the sizeof the web is the light ldquodotrdquo in the upper left corner (DSC_3146) (F) The closely woven horizontal orb-web of an undescribed Tetragnathidaefrom Australia (DSC_8075) (G) The horizontal sheet-web of an undescribed Linyphiidae from Australia (DSC_2794) (H) Detail of (G)(DSC_2801) Photos G Hormiga
234 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
had already been transformed into significantly differ-ent architectures such as those found in linyphioids(sheet-webs) and theridiids (cob- and sheet-webs) Theancestors of the RTA clademdasha lineage that includesmany ground and cursorial spiders such as wolf(Lycosidae) and jumping spiders (Salticidae)mdashmayhave built orb-webs Throughout their diversificationorb-weavers have often abandoned foraging webs toadopt a cursorial lifestyle (eg Fig 8A B C F) Inde-pendent and well-supported cases of araneoids thathave abandoned ancestral foraging snares in favour ofactive hunting for prey include the oarcine araneids(eg Oarces sp Fig 8B) the leaf-litter inhabiting fam-ily Malkaridae (Figs 8F 9AndashC) Mimetidae (a largelyaraneophagic lineage Fig 8C) the arkyids (which wenow classify in the family Arkyidae Fig 8A) and theholarchaeids (which we now classify in the familyAnapidae Fig 9E F) There are some striking conver-gent morphological features associated with some ofthese independent instances of evolution of cursorialforaging behaviour such as the leg spination patternof mimetids (Fig 8C) New Zealand malkarids(Fig 10H) and of some of the oarcine araneids
(Fig 8B) in which the anterior leg or legs share anarrangement of macrosetae alternating distinctivelylong and short spiniform setaeOrbs are old (Late Triassic to early Jurassic Fig 4)
and likely have a single origin (eg Bond et al 2014Fernandez et al 2014) but the RTA clade taxa haveeither abandoned building orb-webs or have shifted todifferent web architectural types such as the sheet-webs of agelenids or the irregular ground-webs ofamaurobiids It seems now that from a systematicpoint of view the orb-web itself is not a good charac-ter (or character complex) with which to define cladesThus a logical consequence of these results (see alsoBond et al 2014 Fernandez et al 2014) is to aban-don the concepts of Orbiculariae (Araneoidea plusDeinopoidea) and Deinopoidea (Deinopidae plus Ulo-boridae) because neither of them correspond to mono-phyletic groups orbicularian could still be used in thevernacular sense but not to refer to a taxon or a natu-ral groupSimilarly to web architecture web type (cribellate or
ecribellate) has also had a very dynamic evolutionaryhistory However it has been dominated by a general
(A) (B)
(C) (D)
(E) (F)
Fig 7 Webs of Physoglenidae (A) Physoglenes sp from Chile (GH001230_R03_14) (B) Mangua sp from New Zealand (DSC_7925) (C)Chileotaxus sp from Chile (DSC_2028) (D) Undescribed physoglenid from Australia (DSC_1392) (E) Pahora parakaunui from New Zealand(CASENT9062577_CRW_0363) (F) Runga sp from New Zealand (DSC_7972) Photos G Hormiga except (E) (C Griswold)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 235
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
trend of loss of the cribellum and shift to eitherecribellate webs or cursorial (non web-building) life-styles As in previous analyses when a model of char-acter transformations with equal rates is consideredthe data are best explained by multiple independentorigins of the cribellum and the cribellate web This ishowever highly unlikely as already argued (eg Milleret al 2010) Nevertheless the use of models thatallow for asymmetric rates of character transforma-tions provides strong support for the single origin ofthe cribellum in agreement with the current view oncribellate web evolution
Systematics of Araneoidea and Nicodamoidea
In this section we discuss the taxonomic and system-atic implications for Araneoidea based on the phyloge-netic results of this study (as well as data presentedelsewhere) Membership and composition of higher-level groups are discussed for extant taxa only We
have chosen the results of the ML analyses of the fulldata matrix to guide our taxonomic decisions (Figs 2and S3) but the taxonomic decisions take into accountthe results from other methods degrees of supportand morphological characters that aid the diagnoses ofgroups discussed hereBased on the phylogenetic results of this study the
superfamily Araneoidea includes the following 17 fam-ilies Anapidae Araneidae Arkyidae CyatholipidaeLinyphiidae Malkaridae Mimetidae MysmenidaeNesticidae Physoglenidae Pimoidae Symphytognathi-dae Synaphridae Synotaxidae TetragnathidaeTheridiidae and Theridiosomatidae Micropholcom-matines constitute a lineage within Anapidae The lat-ter would be rendered paraphyletic if the former weretreated at the family rank as demonstrated byLopardo et al (2011) (see also Lopardo and Hormiga2015 and discussion below)We highlight the following higher-level taxonomic
changes that are discussed in more detail below
(A)
(B)
(C)
(D)
(E)
(F)
Fig 8 (A) Arkys sp (Arkyidae) a web-less araneoid from Australia (DSC_0191) (B) Oarces sp (Araneidae) a web-less araneoid from Chile(DSC_2399) (C) The pirate spider Gelanor latus (Mimetidae) from Brazil (DSC_9119) (D) The cribellate Megadictyna thilenii (Megadictynidae)from New Zealand (DSC_2599) (E) An Australian member of the ecribellate family Nicodamidae (DSC_2729) (F) An undescribed cursorialspecies of Malkara (Malkaridae MALK_GH_017) from Australia (DSC_8196) Photos G Hormiga
236 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
The cribellate and ecribellate nicodamids are nowranked at the family level (Megadictynidae rank res
and Nicodamidae stat n respectively) and groupedunder the superfamily Nicodamoidea rank n Synotaxi-dae are now circumscribed to include only the genusSynotaxus The formerly synotaxid subfamiliesPhysogleninae and Pahorinae are now grouped underthe family Physoglenidae rank n Arkyinae formerly inAraneidae is now classified as the family Arkyidaerank n Nephilinae rank res is now classified as a sub-family under the re-circumscribed family Araneidae
The results also corroborate the placement of Oarcinaein Araneidae rather than in Mimetidae as formallyproposed by Dimitrov et al (2012) The morphologyof Sinopimoa bicolor the only member of the familySinopimoidae (Li and Wunderlich 2008) as describedso far is congruent with that of Linyphiidae (Hor-miga 2008) and thus we consider Sinopimoidae ajunior synonym of the family Linyphiidae (syn n)Holarchaeidae is a junior synonym of the familyAnapidae (syn n) and Pararchaeidae a junior syn-onym of the family Malkaridae (syn n)
(A)
(C) (D)
(F)
(B) (E)
Fig 9 (A B) A female of the Tasmanian malkarid Ozarchaea ornata (Malkaridae formerly Pararchaeidae) dorsal (A) ventral (B) (C D) Themale of an undescribed species of Malkara (Malkaridae MALK_GH_013) from Australia dorsal (C) ventral (D) (E) Lateral view of the ante-rior region of the prosoma of a female of Holarchaea (Anapidae) from New Zealand showing its highly modified chelicerae (F) A male ofHolarchaea (Anapidae) from New Zealand dorsal Photos G Hormiga (E F Griswold lab-ATOL project)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 237
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
(A) (B) (C)
(D)
(E)
(F) (G)
(H)
Fig 10 (A B) SEM of the male pedipalp (right reversed) of Pararchaea sp (Malkaridae) from Australia ectal (A) ventral (B) (C) SEM of themale pedipalp (left) of an undescribed Malkaridae (MALK_GH_009) from New Zealand ventral (D E) Male of Pararchaea sp (Malkaridae)from Australia dorsal (D) anterior with open chelicerae (E) (F) Female of Pararchaea sp (Malkaridae) from Australia anterior The cheliceralpeg teeth can be seen next to the fangs (G) SEM of the male tarsal organ of Holarchaea (Anapidae) from New Zealand (H) SEM of the femur Ispination pattern of an undescribed Malkaridae (MALK_GH_009) from New Zealand C Conductor CA Conductor Apex CBA ConductorBasal Apophysis E Embolus EB Embolus Base T Tegulum P Paracymbium Photos G Hormiga (A B G Griswold lab-ATOL project)
238 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Taxonomy
Araneae Clerck 1757
Superfamily Nicodamoidea Simon 1897 rank n
Diagnosis (after Harvey (1995) and Griswold et al(2005)) male palpal tibia with large dorsal apophysistarsi without trichobothria Cribellate nicodamoids dif-fer from Phyxelididae in lacking a clasping spine onmale metatarsus I and lacking thorn-like setae on theanterior of the palpal femora They differ from Tita-noecidae in having a simple dorsal tibial apophysis onthe male palp and having paracribellar spigots on thePMSPutative synapomorphies dorsal tibial apophysis in
the male palp (Harvey 1995 Griswold et al 2005Ramırez 2014) the complex conformation of this pro-cess (Ramırez 2014 p 241) branched median tra-cheae (Griswold et al 2005) and a single cheliceraltooth (Harvey 1995) have been suggested as providingmorphological evidence of Nicodamoidea monophylyComposition Two families Nicodamidae Simon
1958 3101Type species Theridion peregrinum Walckenaer
1841 297 = Nicodamus peregrinus (Walckenaer 1841)Diagnosis (based in part on Harvey (1995)) Ecribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig172AndashD) and a row of three to four stiff dark setae inan otherwise large bare area on the dorsal surface ofthe ALS (Griswold et al 2005 fig 41A C) (Fig 8E)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon loss of thecribellum a row of three to four stiff dark setae in anotherwise large bare area on the dorsal surface of theALS bright red carapace legs and sternum fertiliza-tion duct openings facing mesallyComposition Seven genera with 27 species found in
Australia and New Guinea Included are Ambico-damus Harvey 1995 Dimidamus Harvey 1995 Duro-damus Harvey 1995 Litodamus Harvey 1995Nicodamus Simon 1887 Novodamus Harvey 1995and Oncodamus Harvey 1995
Family Megadictynidae Lehtinen 1967 rank res
Megadictynidae Lehtinen 1967 247 296 Synony-mized with Nicodamidae by Forster 1970 177Type species Megadictyna thilenii Dahl 1906 62Diagnosis (based in part on (Harvey 1995)) Cribel-
late entelegyne spiders with a large dorsal apophysison the male palpal tibia (Griswold et al 2005 fig171AndashC) entire cribellum (Griswold et al 2005 fig
41A B) a posterior mAP spigot on the PLS (Griswoldet al 2005 fig 39C) and enlarged spinning field ofthe PLS (Forster 1970 fig 523 Griswold et al 2005figs 39A D 40A D) (Fig 8D)Putative synapomorphies Harvey (1995) suggests the
following synapomorphies for this taxon the enlargedspinning field of the posterior lateral spinneret and thelocation of the copulatory duct openings onto the dor-sal surface of the epigynumComposition Two genera with two species found in
New Zealand Forstertyna Harvey 1995 and Megadic-tyna Dahl 1906Comments The superfamily Nicodamoidea sister
group to the Araneoidea is readily diagnosed but thesame can be said for each included family We proposethat two families be recognized here resurrecting thestatus of both Megadictynidae and Nicodamidae Theassociation of the cribellate Megadictyna with theecribellate Nicodamidae was first proposed by RayForster based on a suggestion by C L Wilton (For-ster 1970 p 177) This taxonomic grouping was cor-roborated by Harvey (1995) Griswold et al (2005)Blackledge et al (2009) Dimitrov et al (2012 2013)Ramırez (2014) and by this study Nevertheless theconventional Nicodamidae sensu Forster (1970) areheterogeneous Synapomorphic and diagnostic charac-ters of Megadictynidae and Nicodamidae respectivelyserve grouping functions and justify the recognition oftwo familiesThe ecribellate nicodamids had long been associated
with Araneoidea perhaps because of their somaticsimilarity to theridiids (eg Fig 8E) and indeedecribellate nicodamids were attributed originally to thecomb-footed spiders The first described was Theridionperegrinum Walckenaer (1841) from lsquoBrazilrsquo shortlythereafter L Koch (1865) named three others fromAustralia including Theridium semijlavum from Wol-longong New South Wales Although Simon (1898)suggested that Nicodamus was not a theridiid andplaced this genus in the subfamily Nicodaminae inAgelenidae (Simon 1897) Nicodamus continued to becatalogued under Theridiidae (Roewer 1942 Bonnet1958) Herbert and Lorna Levi world experts onTheridiidae rejected theridiid placement for Nico-damus and after discussing the issue with Forster(Forster 1970 p 177) moved Nicodamus to Zodari-idae (Levi and Levi 1962) thereby ending their associ-ation with theridiids and more broadly AraneoideaThe cribellate Megadictyna was described in Dic-
tynidae by Dahl (1906) which placement was followedby Marples (1959) Lehtinen (1967) thought Megadic-tyna so distinct from dictynids and from other spidersthat he created the family MegadictynidaeHarvey (1995) revised Nicodamidae and followed
Forster (1970) by including cribellate and ecribellatemembers providing a diagnosis and suggesting as
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 239
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
synapomorphies the male palpal tibia with large dorsalapophysis metatarsus IV without a trichobothriumand the chelicera with a single distal tooth on the pro-margin Harvey (1995) placed the nicodamids in theldquoRTA claderdquo (ie spiders with any process on the malepalpal tibia) and further could only suggest placementin the ldquoAmaurobioideardquo RTA clade spiders with sim-ple entire or weakly branched tracheal systemsSuggested orb-weaver affinities for Nicodamidae
began to appear a few years later in one of theequally most parsimonious trees for Entelegynae sug-gested by Griswold et al (1999 p 60) Nicodamidaeand Orbiculariae appeared as sister groups althoughthis result was based in part on character codings(eg serrate accessory setae on the tarsi) that werelater discovered to be more widespread orbicularianaffinities of Nicodamidae appeared again in thecladistic analyses of Griswold et al (2005 figs 218BC) Morphological evidence for this arrangementremains weak like Araneoidea Megadictyna have theminor ampullate gland spigot (mAP) on the posteriormedian spinnerets (PMS) posterior (Griswold et al2005 fig 140C) but in ecribellate nicodamids thePMS mAP is median (not anterior nor posterior) andtherefore not informative Placement of nicodamidsoutside the RTA-clade saves some evolutionary stepsthe cribellum of Megadictyna is entire like uloboridsand deinopids and different to most RTA-clade spi-ders and the palpal tibial apophysis is dorsal notretrolateral Nevertheless the morphological evidencefor placing nicodamids near or far from orb-weaversis not robust It is molecular evidence albeit from thesame genes but with a diverse array of taxon samplesthat strongly associates Nicodamoidea with Arane-oidea (Blackledge et al 2009 Miller et al 2010Spagna et al 2010 Dimitrov et al 2012 2013Agnarsson et al 2013) although Nicodamoidea wascontradicted by Agnarsson et al (2012) That result iscorroborated by our analysis with relatively good(73) bootstrap support and we consider this the bestsupported working hypothesis This implies a notablecourse of web evolution from the primitive homolo-gous orb of deinopoids and araneoids to a substrate-limited sheet of cribellate nicodamids unrecognizablearchitecturally as an orb The evolution of the wholeRTA clade from an orbicularian ancestor is thus con-ceivable an idea that has been recently corroboratedby phylogenomic data (Bond et al 2014 Fernandezet al 2014)
Superfamily Araneoidea Clerck 1757
Family Anapidae Simon 1895
Type species Amazula hetschkii Keyserling 1886Micropholcommatidae Hickman 1944 (implied but
not formalized in Brignoli (1970) and Scheuroutt (2003)synonymy formally proposed in Lopardo et al (2011)see also Lopardo and Hormiga (2015))
Type species Micropholcomma caeligenum Crosbyand Bishop 1927Holarchaeidae Forster and Platnick 1984 syn n
Type species Archaea novaeseelandiae Forster 1949Diagnosis Minute Araneoidea with the labium fused
to the sternum a huge posterior PLS cylindrical glandspigot pore-bearing prosomal depressions on the lat-eral margin of the carapace (except most microphol-commatines which do not have pores) and abdomenwith conspicuous sigilla and provided with scatteredsclerotized spotsPutative synapomorphies Anapid synapomorphies
comprise at least the labium fused to the sternum thecarapace with pore-bearing prosomal depressions (lostin most micropholcommatines) and fatiscent leg cuti-cle Additional morphological synapomorphies are dis-cussed and illustrated in Lopardo et al (2011) andLopardo and Hormiga (2015)Composition Fifty-eight genera and 238 species
worldwide Of these 19 genera and 66 species areplaced in Micropholcommatinae and found in SouthAfrica South America Australia and New Zealandand one genus with two species in Holarchaea occur-ring in Australia and New Zealand Many more spe-cies remain to be discovered especially in the tropicsComments The family-level taxa treated here as syn-
onyms have had a convoluted and troubled history Rixand Harvey (2010a p 13) pointed out that ldquoAnapidaeare at the center of all problems lsquosymphytognathi-danrsquo in naturerdquo Micropholcommatidae were long asso-ciated with Araneoidea but in 1984 along withMimetidae and the newly created family Holarchaeidaethey were placed far away in the Palpimanoidea (For-ster and Platnick 1984) The study of Griswold et al(1998) did not address the PalpimanoideaAraneoideaproblem explicitly and treated Araneoidea circumscrip-tion as firmly established (the symphytognathoid fami-lies were included but not the Mimetidae) Scheuroutt (20002003) placed Micropholcommatidae and Mimetidaeback among the araneoids and suggested thatMicropholcommatidae should be synonymized underAnapidae In spite of her clear argumentation herresults were not widely accepted More recently severalstudies some of which included molecular data(Lopardo and Hormiga 2008 2015 Rix et al 2008Rix and Harvey 2010a Lopardo et al 2011) havefirmly placed micropholcommatines within Araneoideaand Wood et al (2012 2013) definitively distinguishedPalpimanoidea and AraneoideaThe status of Micropholcommatidae remained
unsettled with Lopardo and Hormiga (2008) agreeingwith Scheuroutt (2000) in synonymizing them with Anapi-dae Rix and Harvey (2010ab) rejecting this syn-onymy Lopardo et al (2011) reasserting thesynonymy on the basis of a new suit of synapomor-phies and Lopardo and Hormiga (2015) corroborating
240 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
this The placement of Micropholcommatidae as asubgroup of Anapidae can now be considered to bestrongly corroboratedThe family Holarchaeidae (Fig 9E F) is another
story Despite a striking superficial resemblance to thepalpimanoid ldquopelican spidersrdquo (Archaeidae) placingHolarchaeidae in the Palpimanoidea presents a num-ber of problemsmdashsuch as their entelegyne female geni-talia the absence of cheliceral peg teeth and the lackof leg I scopulae Our molecular analysis groupsHolarchaea with the anapid Acrobleps with strong sup-port in all data treatments and in turn these taxagroup with other Anapidae including the type genusAnapis albeit with low support What the moleculardata suggest is strongly corroborated by morphologyLopardo et al (2011) and Lopardo and Hormiga(2015) suggest a number of morphological synapomor-phies for Anapidae and Holarchaea shares most ofthese The labium is fused to the sternum carapacewith pore-bearing prosomal depressions including alarge depression near the carapace lateral margin ster-nal cuticle is punctate leg cuticle is fatiscent the tarsalorgan opening is huge subequal or larger than setalsockets (Fig 10G) abdomen with conspicuous sigillaand it is also provided with scattered sclerotized spotsanterior respiratory system comprises modified book-lungs females have internal copulatory openings sper-matic duct simple with no loops before entering theembolus and thick embolus Like Symphytognathidaemales lack epiandrous fusules and the posterior PLScylindrical gland spigot is enlarged whereas Lopardoand Hormiga (2015) regard these as anapid plus sym-phytognathid synapomorphies on our tree they mayoptimize as anapid synapomorphies Lastly theabsence of a paracymbium from the male palp hasalso been interpreted as an anapid plus symphytog-nathid synapomorphy (Lopardo et al 2011) Never-theless Anapidae continue to be problematic (Rix andHarvey 2010a p 124) because the family optimizes asdiphyletic true Anapidae include Anapis microphol-commatines and the holarchaeids but a second ldquoana-pidrdquo clade comprising Gertschanapis Maxanapis andChasmocephalon resolves elsewhere Only in the parsi-mony analyses are these two anapid clades recoveredas sister groups albeit with low support (Fig S7)Understanding anapid phylogenetic relationships isessential to study evolutionary transitions betweenorb-webs and other architectures Most Anapidaebuild micro-orbs (eg Fig 6B see also Miller et al2009) but the family also includes species that buildsheet-webs similar to those of Cyatholipidae (Hormigaunpublished)
Family Synotaxidae Simon 1894
Synotaxeae Simon 1894 494Synotaxidae Forster Platnick and Coddington1990
Type genus Synotaxus Simon 1895Diagnosis Diagnostic characters for Synotaxidae
(circumscribed here to include only the genus Syno-taxus) include the unique ldquochicken-wirerdquo web compris-ing modular rectangles of sticky silk (Fig 6C) thefollowing character combination further distinguishessynotaxids spiniform setae on the male palpal patella(though at least S ecuadorensis is depicted as havingspiniform setae on the tibia instead (Exline and Levi1965 figs 25ndash27 Griswold et al 1998 fig 19C)enlarged (but not flattened) aggregate gland spigots onthe PLS (Griswold et al 1998 figs 38A D) legfemora not basally thickened a retrolateral groove onthe paracymbium and a dorsally-excavated and cup-shaped integral paracymbium (Griswold et al 1998fig 19C Agnarsson 2004a fig 3)Putative synapomorphies The unique ldquochicken-wirerdquo
web comprising modular rectangles of sticky silk(Eberhard 1977 1995) other homoplastic synapo-morphies comprise spiniform setae on the male palpalpatella (shared with some Physoglenidae eg Nomauacrinifrons) enlarged (but not flattened) aggregate glandspigots on the PLS a retrolateral groove on the para-cymbium (shared with Physoglenidae) and a dorsally-excavated and cup-shaped integral paracymbium(shared with Cyatholipidae and Physoglenidae)Composition Only the genus Synotaxus with 10 spe-
cies endemic to the American tropicsComments Forster et al (1990) associated Syno-
taxus with Physoglenes Pahora and other similar gen-era in the new family-ranked Synotaxidae Wedistinguish Synotaxidae and Physoglenidae as separatefamilies to recognize the separate affinities on our treeand to make each family easier to diagnose Such dif-ferences in genealogical relationships help to explainthe great disparity in web architecture between syno-taxids (vertical ldquochicken-wirerdquo modular webs Fig 6C)and the physoglenids (horizontal sheet and irregularwebs Fig 7) In addition the different geographicaldistribution of these two groups better fits the currentphylogenetic re-circumscription
Family Physoglenidae Petrunkevitch 1928 rank n
Type Genus Physoglenes Simon 1904Diagnosis Physoglenids have lost the basal PLS
cylindrical spigot and any cylindrical spigots from thePMS (Griswold et al 1998 figs 40 42 44) likeSynotaxidae they have a retrolateral cymbial incisionand like Synotaxidae and Cyatholipidae they have asmall basal dorsally-excavated paracymbium (Gris-wold et al 1998 figs 18CndashF) Physoglenids differfrom Cyatholipidae in having the posterior trachealspiracle narrower than the width of the spinneretsMembers of subfamilies Physogleninae and Pahorinaehave modifications of the male abdomen and cara-pace andor abdomen that may function in stridula-tion
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 241
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Putative synapomorphies The loss of the cylindricalgland spigots from the PMS is a unique synapomor-phy homoplastic synapomorphies include the para-cymbium and cymbial form elongate but basallythickened femora truncate posterior apex of the ster-num and complex tegular apophysis which may behomologous either to the conductor (Griswold et al1998) or the theridiid tegular apophysis (Agnarsson2004b)Composition Thirteen genera and 72 species found
in Australia New Zealand and southern South Amer-ica (Argentina and Chile) additional genera and spe-cies remain to be describedComments Synotaxus and genera here newly
assigned to the Physoglenidae were associated in theSynotaxidae by Forster et al (1990) They suggestedthat potential synapomorphies were the small basaldorsally-excavated paracymbium a retrolateral cym-bial incision dorsal macrosetae on the male palp(though the segment varies and some lack such setaealtogether) and greatly elongated spineless legs (For-ster et al 1990) Our analyses consistently separateSynotaxus from other former members of Synotaxidaealthough support values for the intervening nodes arelow Nevertheless we recognize Physoglenidae andSynotaxidae as separate families The monophyly ofPhysoglenidae in our analysis (Pahora Runga Mer-inga Tupua Physoglenes Mangua Chileotaxus andSynotaxidae sp (GH1194) an undescribed genus fromNew Zealand) receives maximum clade support Phy-soglenids are sister group to the pimoidlinyphiid lin-eage albeit with a low support value As discussedabove Synotaxus appears elsewhere in our tree dis-tantly related to physoglenids Recognizing Physogle-nidae and Synotaxidae as separate families iscognizant of these separate phylogenetic affinities andmakes each family easier to diagnose A diagnosticcharacter for the Physoglenidae is the absence of anycylindrical gland spigots from the PMS Other poten-tial physoglenid synapomorphies are shared with otherfamilies only a single cylindrical gland spigot remain-ing on the PLS (shared with Cyatholipidae) retrolat-eral groove on the paracymbium (shared withSynotaxidae) and dorsally-excavated cup-shaped inte-gral paracymbium (shared with Cyatholipidae andSynotaxidae) Dorsal macrosetae or cuticular spurs onthe male palp are not universal and may characterizegenera or subgroups of Physoglenidae Most physogle-nid genera have some form of carapaceabdomenstridulating mechanism although nothing of the sortis found in Chileotaxus which nevertheless agrees withthe other Physoglenidae in the PMS and PLS spinneretsynapomorphies In addition to explaining the differ-ences in web architecture between synotaxids (Fig 6C)and physoglenids (Fig 7AndashF) our phylogenetichypothesis also helps to explain the similarities in the
sheet-webs of some physoglenids and some linyphiidsFor example the sheet-web of the Chilean Physoglenespuyehue (Fig 7A) could easily be taken as a linyphiidweb (Fig 6G)
Subfamily Physogleninae Petrunkevitch 1928
Type Genus Physoglenes Simon 1904Diagnosis The anterior part of the abdomen of phy-
soglenine males is sclerotized in association with anexpanded heavily sclerotized pedicel (Forster et al1990)Composition Five genera and 20 species Included
are Physoglenes Simon 1904 from South AmericaMeringa Forster 1990 and Zeatupua Fitzgerald andSirvid 2009 from New Zealand and Tupua Platnick1990 and Paratupua Platnick 1990 from Australia
Subfamily Pahorinae Forster 1990 (in Forster et al
1990 36)
Type Genus Pahora Forster 1990 (in Forster et al1990 40)Diagnosis Forster et al (1990) suggest that pahori-
nes can be diagnosed by an area on the posterior mar-gin of the carapace that engages with a stridulatoryfile on the antero-dorsal surface of the abdomen ofmalesComposition Four genera and 34 species all from
New Zealand Included are Pahora Forster 1990Pahoroides Forster 1990 Nomaua Forster 1990 (asenior synonym of Wairua Forster 1990 see (Fitzger-ald and Sirvid 2009)) and Runga Forster 1990Comments There are two unplaced physoglenid gen-
era from New Zealand (Mangua Forster 1990 and anew genus discussed below) one (Chileotaxus Plat-nick 1990) from South America and two (Calcarsyno-taxus Wunderlich 1995 and MicrosynotaxusWunderlich 2008) from Australia All of these generalack the peculiar carapaceabdomen modifications forstridulation that are found in Pahorinae andPhysogleninae Chileotaxus and Mangua have the pal-pal and spinneret modifications characteristic of Phy-soglenidae Chileotaxus is sister group to Physoglenesin our analysis with high support value and Manguagroups with these two genera with lower support Anundescribed New Zealand physoglenid (Synotaxidaesp [GH1194]) has been found as either a commensalor a kleptoparasite in the webs of cyatholipids (For-ster 1988 pp 8ndash9 Forster and Forster 1999 p 195Paquin et al 2010 p 61) stiphidiids and hexathelids(CG and GH pers obs) This small (2 mm) spiderwith a round abdomen and enlarged divergent malechelicerae closely resembles cyatholipids in the genusTekella in whose webs they may live In contrast tocyatholipids the hexathelids and stiphidiids and thehost sheet-webs in which these undescribed physogle-nids live are both significantly larger than the com-mensalkleptoparasites In every mention they havebeen identified as theridiids but their palpal form
242 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
especially the small cup-shaped paracymbium placesthem in Physoglenidae In our analysis these groupwith the Pahorinae genera Runga and Pahora with aBS = 72 The Australian genera Calcarsynotaxus andMicrosynotaxus are of dubious affinities Calcarsyno-taxus has only one PLS cylindrical gland spigot (likePhysoglenidae) a small basal dorsally-excavatedparacymbium and a pair of strong male palpal patellaspiniform setae (like Synotaxidae and many Physogle-nidae) like Synotaxidae Calcarsynotaxus has a cylin-drical gland spigot on the PMS Microsynotaxus lacksa PMS cylindrical and has only one PLS cylindricalbut the male palp is unlike any Synotaxidae or Physo-glenidae Based on our phylogenetic hypothesis thePahorinae are monophyletic (and include the unde-scribed genus from New Zealand) and the Physogleni-nae as currently circumscribed which may includeChileotaxus and Mangua are paraphyletic Additionaltaxa need to be added to the analysis (especially Cal-carsynotaxus and Microsynotaxus) before taking fur-ther taxonomic actions
Family Malkaridae Davies 1980 stat n
Type genus Malkara Davies 1980Type species Malkara loricata Davies 1980Family Pararchaeidae Forster and Platnick 1984syn n
Type species Pararchaea alba Forster 1955Diagnosis Small to very small cryptic entelegyne three
clawed spiders Male palp with basal paracymbium nomedian apophysis and a conductor that circles theembolus in opposite direction to most araneoids (coun-terclockwise left palp ventral view Fig 10B C) Bodyarmored with a ventral abdominal scutum around thepedicel in males (sometimes also in females) and sclero-tized ring around spinnerets in both sexes Abdomenwith sclerotized sigilla (Figs 9A C 10D) Like mime-tids both sexes lack aggregate and flagelliform glandspigots on posterior lateral spinnerets Some malkaridsparticularly some of the New Zealand species have leg Iand II spination very similar to that of mimetids (alter-nating long and short spines Fig 10H) but malkaridscan be distinguished from the latter by the unique orien-tation of the palpal conductor (Fig 10C)Putative synapomorphies Abdomen with ventral
abdominal scutum that surrounds pedicel (at least inmales) sclerotized ring around spinnerets (Fig 9D)abdominal setae arise from sclerotized discs (Fig 9AC) abdomen with sigilla (Fig 9A C) sternum fusedaround petiole to carapace conductor encircling theembolus in a counterclockwise direction and with aconspicuous basal apophysis (Fig 10C) PLS araneoidtriad absent (Rix and Harvey 2010b figs 16ndash17)Composition Eleven genera and 46 described spe-
cies Included are the genera Anarchaea CaratheaChilenodes Flavarchaea Forstrarchaea Malkara Nan-archaea Ozarchaea Pararchaea Perissopmeros and
Westrarchaea Numerous new malkarid species remainto be described from New Zealand (at least 12 newspecies) and Australia (Hormiga and Scharff unpub-lished)Comments The spider family Pararchaeidae was
erected by Forster and Platnick (1984) to accommodatefive Australian and two New Zealand Pararchaea spe-cies described by Forster (1955) Forster and Platnickplaced this family within the superfamily Palpi-manoidea together with two other new families estab-lished in the same paper Mecysmaucheniidae andHolarchaeidae Scheuroutt (2000) tested the limits of Palpi-manoidea and Araneoidea in a phylogenetic study andconcluded that Pararchaeidae Holarchaeidae andMimetidae belonged in the superfamily AraneoideaThis placement was confirmed by a molecular study(Rix et al 2008) and the placement of Holarchaeidaeand Pararchaeidae within Araneoidea was further cor-roborated by Wood et al (2012) based on both molec-ular and morphological data In our study we findstrong support for a placement of Pararchaeidae withinthe current family Malkaridae thereby rendering thislatter family paraphyletic and we therefore synonymizePararchaeidae with Malkaridae Some of our analysessupport a sister-group relationship between Pararchaei-dae and Malkaridae If both current families (Malkari-dae and Pararchaeidae) turn out to be reciprocallymonophyletic they could be ranked as subfamilieswhile retaining the family diagnosis that we have pro-vided here for the recircumscribed MalkaridaeOur results support four clades within the re-circum-
scribed Malkaridae (but see the parsimony results) alineage with the representatives of Perissopmeros Car-athea and Chilenodes (ie subfamily Sternoidinae Har-vey 2002) a lineage with the New Zealand taxa (all ofwhich are currently undescribed and including at least12 new species) a lineage with Malkara (currentlymonotypic but there are no less than 30 undescribedspecies in Australia) and a lineage with the formerpararchaeid representatives (see Rix 2006) It is worthmentioning that in the results from the parsimonyanalyses pararchaeids did not cluster with malkaridsbut with a clade containing mostly cyatholipids how-ever this grouping and all intermediate branchesbetween malkarids and that clade did not receive sig-nificant supportThe new expanded Malkaridae consist of species
found mainly in Australia and New Zealand Onlytwo of the 46 known species have been found outsidethis region That is Flavarchaea humboldti Rix andHarvey 2010ab from New Caledonia and Chilenodesaustralis Platnick and Forster 1987 from Argentinaand Chile
Family Arkyidae rank n
Arkyinae L Koch 1872Type genus Arkys Walckenaer 1837
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 243
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Type species Arkys lancearius Walckenaer 1837Diagnosis (mainly from Framenau et al 2010) the
prolateral field of short dense setae on tarsus I ofmales (Heimer et al 1982) and the enlarged aggregategland spigots on the PLS of both sexes are unique toArkyidae Arkyidae can be further diagnosed by thefollowing combination of characters both sexes with aprocurved posterior eye row and with posterior med-ian eyes more widely spaced than the anterior medianeyes absence of radix and abdomen distinctively trian-gular in males (Fig 8A) and a pattern of abdominalsigilla in two rows in females Arkyids are distin-guished from most araneids and tetragnathids by theabsence of foraging websPutative synapomorphies prolateral field of short
dense setae on tarsus I of males (Framenau et al2010 fig 2) enlarged aggregate gland spigots on thePLS (Framenau et al 2010 figs 4A CndashD 22D and23D) in both sexes and absence of a flagelliform glandspigot (Framenau et al 2010 figs 22D and 23D)Composition Two genera (Arkys Walckenaer 1837
and Demadiana Strand 1929) and 37 species (WorldSpider Catalog 2016) v170Comments Eight species of Arkys have been
included in this study a broad and representative sam-ple of the morphological variation within the genusThe genus Demadiana could not be included becauseDNA quality tissue was not available but there arestrong morphological synapomorphies that uniteDemadiana and Arkys (see diagnosis) as a mono-phyletic group (Framenau et al 2010) In this studybased entirely on molecular data there is strong sup-port for the monophyly of the genus Arkys (BS = 100)and strong support for the sister-group relationship toTetragnathidae (BS = 100) Together with MimetidaeTetragnathidae and Arkyidae constitute a mono-phyletic group with high support (BS = 98)The systematic position of Arkys has been contro-
versial Previous authors have placed the genus in suchdifferent families as Thomisidae Araneidae Tetrag-nathidae and Mimetidae (for the taxonomic history ofthis group see Framenau et al 2010) Heimer (1984)placed Arkys in Mimetidae based on the complicatedparacymbium of the male palp and this placementwas supported by Platnick and Shadab (1993) How-ever Platnick and Shadab (1993) reported the presenceof aggregate gland spigots on the posterior lateralspinnerets of Arkys and thereby contradicting themimetid placement aggregate gland spigots are knownonly from Araneoidea and Platnick and Shadab(1993) considered mimetids to be palpimanoids notaraneoids (Forster and Platnick 1984) Scharff andCoddington (1997) tested the monophyly and phyloge-netic placement of Arkys within Araneoidea in a mor-phological matrix and found Arkys to be nested withinAraneidae where until now Arkys has been classified
The molecular analysis of Blackledge et al (2009)found strong support for the placement of Arkys assister group to Tetragnathidae and for a sister-grouprelationship between a clade consisting of Tetragnathi-dae + Arkys and Mimetidae as also found by Dim-itrov et al (2012) and the current study Thecombined analyses of Dimitrov and Hormiga (2011)also refuted araneid affinities of Arkys but could notunambiguously resolve its placement Some analysessuggested that Arkys was sister group to Tetragnathi-dae (all Bayesian analyses as in Blackledge et al2009) whereas in other analyses Arkys appears to be amimetid (dynamic and static homology parsimonyanalyses and the morphological partition)Our analyses as well as the above-cited molecular
analyses place Arkys as the sister group of Tetrag-nathidae with high support values In a guide to theorb-weaving spiders of Australia Davies (1988 p 282)ldquotentatively placed within the metinesrdquo the genusArkys based solely in the absence of mimetidpalpi-manoid characters Davies did not offer any explicitcharacter support for a metainetetragnathid groupingThis is not surprising as no characters had ever beensuggested to justify a circumscription of Tetragnathi-dae that would include Arkys We treat Tetragnathidaeand Arkyidae as separate families thereby fulfillingthe reciprocal monophyly requirement and makingboth families easier to diagnose morphologically
Family Araneidae Clerck 1757
Type Araneus Clerck 1757Type species Araneus angulatus Clerck 1757Subfamily Nephilinae Simon 1894 rank res
Type Nephila Leach 1815Type species Aranea pilipes Fabricius 1793Diagnosis Araneidae are small to very large enteleg-
yne three-clawed spiders that build typical vertical orb-webs above ground Legs spiny clypeus usually lowMale palp typically complex with at least one tegularsclerite (usually the conductor) with an enlargedembolus base (radix) fused to the proximal part of theembolus in nephilines (Kuntner et al 2008) Adultmales often smaller than females and with pear-shapedcarapace Females with chilum and denticles on che-licerae Fourth leg with sustentaculum (Scharff andCoddington 1997 Griswold et al 1998 Kuntneret al 2008)Putative synapomorphies The presence of modified
setae (sustentaculum) on the tip of the fourth tarsi andthe presence of a radix in the embolic division of themale palp are putative synapomorphies of AraneidaeThe radix is fused to the proximal part of the embolusin nephilines and in a few other araneids (eg Neogea)Coddington (1986 pp 339ndash340) suggested the presenceof nonbirefringent cement at all SS-line and radiusjunctions (SS-R cement) as another potential synapo-morphy of araneids
244 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Composition Araneidae excluding Arkyinae (Arkysand Demadiana now family Arkyidae) but includingthe subfamily Nephilinae (Clitaetra Nephila HerenniaNephilengys and Nephilingis) holds 174 genera and3160 species (World Spider Catalog v170 2016)found worldwide many additional species and generaremain to be describedComments Throughout history the family Aranei-
dae has been recognized as a natural group eventhough the taxonomic composition has changed overtime and defining morphological characters have beendifficult to identify The family is diverse morphologi-cally ecologically and behaviorally and this adds tothe difficulties of circumscribing the family The lastcomprehensive classification is that of Simon (1892)His concept of Araneidae (Argiopidae) was more simi-lar to the modern-day superfamily Araneoidea than tomodern-day Araneidae Subsequent attempts to cir-cumscribe the superfamily have been done mainlythrough re-delimitation and redefinition especiallywithin the large families Araneidae and Theridiidae(Coddington and Levi 1991) Until recently the familyAraneidae included present-day Theridiosomatidaeand Tetragnathidae but these families were removedfrom Araneidae thereby making it more compact anddiagnosable For most of the 20th Century and beforeNephila and its relatives were considered as a subfam-ily (Nephilinae) of Araneidae (Simon 1864 1892Roewer 1942 Bonnet 1955 Benoit 1962 Brignoli1983 Heimer and Nentwig 1983 Wunderlich 19862004 see Kuntner et al 2008 for a historical over-view) until Levi (1986) suggested that Nephila andNephilengys belonged in Tetragnathidae based onmale palpal characters The association of nephilidswith tetragnathids was first shown by the cladisticanalysis of Coddington (1990) based on morphologi-cal and behavioral data and further corroborated lateron with more morphological characters and additionaltaxa by Hormiga et al (1995) Griswold et al (1998)and Dimitrov and Hormiga (2009) Nevertheless thesister-group relationship of nephilids and tetragnathidswas refuted on the basis of redefined and new morpho-logical as well as behavioral characters (Kuntneret al 2008 and simplified versions of this matrix infew other earlier publications) These new studies sug-gested that nephilines were not closely related toAraneidae or Tetragnathidae but could not satisfacto-rily resolve the placement of nephilines among arane-oids The analysis of Kuntner et al (2008) suggestedthat nephilines were the sister group of a clade thatincluded all other araneoid taxa sampled although thiswas only weakly supported Kuntner (2006) removednephilines from Tetragnathidae and raised the groupto family rank (Nephilidae) The first molecular studyincluding nephilines is that of Pan et al (2004) whofound in all of their analyses that Nephila was sister
group to the araneid taxa (a clade of two species)rather than to their two Tetragnatha species Theseauthors suggested that nephilines should be movedback into the Araneidae Their results were howeverbased on a sparse taxon sample and few genes (12SrRNA and 18S rRNA and major ampullate spidroin-1 MaSp1 for a total of nine species) and thusrequired further testing with more taxa and genesStudies by Blackledge et al (2009) using six genes and44 genera and by Alvarez-Padilla et al (2009) usingsix genes and 213 morphological and behavioral char-acters coded for 47 genera confirmed the sister-grouprelationship between nephilids and araneids with highsupport values Further analyses combining morpho-logical and behavioral data (Dimitrov and Hormiga2011) molecular data only (Dimitrov et al 2012) andphylogenomic data (Bond et al 2014) also corrobo-rated the araneid affinities of nephilids A more recentanalysis of nephilid relationships (Kuntner et al2013) based on morphological and molecular dataand analyzing the largest sample of nephilid species todate places nephilids within Araneidae Most of theiranalyses offered high support to a clade that includedall the nephilid and araneid taxa (12 representatives)studied Their results rather consistently imply that thesister group of nephilids are either ldquoaraneids sensustricto in fig 1 and ldquozygiellidsrdquo in fig 2rdquo (Kuntneret al 2013 p 972)Our study including 363 taxa and seven genes
strongly supports the monophyly of a group thatincludes nephilids plus araneids Not surprisingly thecombined group is difficult to define morphologicallybut because all recent phylogenetic analyses (seeabove) and this study have found strong support fora monophyletic Araneidae including nephilids wedecided to return the nephilid lineage to its classicalposition as a subfamily (Nephilinae) within AraneidaeWe are currently only able to list a few putative mor-phological synapomorphies to define the re-circum-scribed Araneidae and have therefore given preferenceto the strong molecular support to guide our decisionfor this taxonomic change Araneidae without nephili-nes are also difficult to define morphologically andsuch a group has low support in all analyses Thischange in rank better reflects our improved under-standing of the phylogenetic position and evolutionaryhistory of nephilines while maintaining the diagnos-ability of Nephilinae and avoids the paraphyly ofAraneidae implied by several recent published studies(eg Kuntner et al 2013 but see Gregoric et al2015) and by the Bayesian results of this study(Fig S2)
Family Linyphiidae Blackwall 1859
Type genus Linyphia Latreille 1804Type species Araneus triangularis Clerck 1757Sinopimoidae Li and Wunderlich 2008 syn n
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 245
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Type species Sinopimoa bicolor Li and Wunderlich2008Comments Although we could not include in our
analysis Sinopimoa bicolor Li and Wunderlich 2008the sole member of Sinopimoidae we formalize herethe hypothesis of Hormiga (2008) stating that Sinopi-moa is a member of the family Linyphiidae Asdetailed by Hormiga (2008 p 4) the study of Li andWunderlich (2008) is missing essential morphologicaldata for a convincing phylogenetic justification of anew family As those authors point out two characterssupport membership of Sinopimoa in the ldquolinyphioidrdquoclade (Pimoidae + Linyphiidae) cheliceral stridulatorystriae and patella-tibia leg autospasy The apparentabsence of conductor and median apophysis in themale palp (one of these sclerites or both are found inPimoidae) supports the conjecture that Sinopimoabicolor is a linyphiid Sinopimoa as described by Liand Wunderlich (2008) does not have any pimoidsynapomorphies In addition Sinopimoa shares twoErigoninae synapomorphies (Hormiga 2000 Millerand Hormiga 2004) absence of the female palpal clawand a retrolateral tibial apophysis in the male palpand like many erigonines is of very small sizeand has only one dorsal tibial spine in legs III and IVThe most parsimonious interpretation of the avail-able data is that Sinopimoa is a linyphiid and conse-quently we treat Sinopimoidae as a junior synonym ofLinyphiidaeOur results suggest with high support that the
pimoid species Weintrauboa and Putaoa group withthe linyphiid genus Stemonyphantes The monophylyof Linyphiidae including the latter clade is alsorobustly supported in our ML and Bayesian results(but see results from parsimony analyses for alterna-tive topology Fig S7) Pimoa plus Nanoa are the sis-ter group of such a Linyphiidae circumscription Theseresults could support a transfer of Weintrauboa andPutaoa to Linyphiidae as members of the subfamilyStemonyphantinae (which would need a significantrevision of its morphological diagnosis) and re-cir-cumscribe Pimoidae to include only Pimoa and NanoaSuch a hypothesis is in conflict with the results of mor-phological analyses (eg Hormiga 2008 Hormiga andTu 2008) A recent interpretation of the male palpsclerites of Stemonyphantes (Gavish-Regev et al 2013)suggests that in this linyphiid genus the tegular scle-rites could be homologues of the conductor and med-ian apophysis but supported a sister grouprelationship of Weintrauboa and Pimoa (and thusPimoidae monophyly) and of Pimoidae plus Linyphi-idae We are currently studying the phylogeny ofpimoids with additional morphological and moleculardata and a much larger taxon sample including unde-scribed taxa (Hormiga and Dimitrov unpublished)Preliminary analyses of the combined and molecular
data robustly support pimoid monophyly includingWeintrauboa and Putaoa and linyphiid and linyphioidmonophyly (Hormiga and Dimitrov 2010) We willaddress this problem with a more extensive data setelsewhere
Acknowledgements
We thank Martın Ramırez and Michael Rix fortheir helpful comments on an earlier draft of themanuscript Field and museum work in Australia wasmade possible through the generous help and supportof Robert Raven Barbara Baehr Mark HarveyMichael Rix L Joy Boutin Martın Ramırez JamieSeymour Diana Silva Davila Catherine Byrne SimonGrove Kirrily Moore and Graham Milledge In NewZealand we were also greatly helped by Cor VinkJagoba Malumbres-Olarte Grace Hall Phil SirvidTeresa Meikle Hannah Wood Diana Silva Davilaand Barry Fitzgerald CG was helped in Chile by Eliz-abeth Arias Liz Morrill Lina Almeida Silva and Han-nah Wood and in South Africa by Teresa MeikleCharles Haddad Ester van der Westhuisen LinaAlmeida Silva and Hannah Wood Jan Pedersenassisted us in the fieldwork in Australia and New Zeal-and Laura Garcıa de Mendoza helped us with speci-men curation We thank Norman Platnick (AmericanMuseum of Natural History) and Petra Sierwald (FieldMuseum of Natural History) for specimen loans ErinMcIntyre assisted with DNA sequencing We thankthe Willi Hennig Society for subsidizing and makingthe program TNT freely available This research wassupported by US National Science Foundation grantsDEB 1144492 114417 ldquoCollaborative ResearchARTS Taxonomy and systematics of selectedNeotropical clades of arachnidsrdquo and 14573001457539 ldquoCollaborative Proposal Phylogeny anddiversification of the orb weaving spiders (Orbicular-iae Araneae)rdquo to GH and GG GHrsquos work at theUniversity of Copenhagen (Zoological Museum) wassupported by a scholarship from Danmarks National-bank and a joint grant with NS from the CarlsbergFoundation MArsquos stay as visiting scholar at HarvardUniversity was funded by the Ministerio de EducacionCultura y Deportes of Spain (PRX1500403) LBacknowledges support from a Weintraub Fellowshipfrom the Department of Biology at GWU and by afellowship from COLCIENCIAS (DepartamentoAdministrativo de Ciencia Tecnologıa e InnovacionDoctorados en el exterior 646) Additional supportcame from The Exline-Frizzell Fund of the CaliforniaAcademy of Sciences and grants from the SchlingerFoundation to CG CG acknowledges NSF supportfrom DEB-0072713 ldquoTerrestrial Arthropod Inventoryof Madagascarrdquo and DEB 9296271 ldquoSystematics and
246 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Biogeography of Afromontane Spidersrdquo CEG andGH also acknowledge support form NSF grants DEB-0613775 ldquoPBI Collaborative Research The Megadi-verse Microdistributed Spider Family Oonopidaerdquoand EAR-0228699 ldquoAssembling the Tree of Life Phy-logeny of Spidersrdquo NS acknowledges the DanishNational Research Foundation for support to the Cen-ter for Macroecology Evolution and Climate Fundingfor this research was also provided by grants from theDanish Agency for Science Technology and Innova-tion (project 272-08-0480) and the Carlsberg Founda-tion (grants 2008-01-0362 2010-01-0185 2010-01-0186)to NS
References
Agnarsson I 2004a The phylogenetic placement andcircumscription of the genus Synotaxus (Araneae Synotaxidae)a new species from Guyana and notes on theridioid phylogenyInvertebr Syst 17 719ndash734
Agnarsson I 2004b Morphological phylogeny of cobweb spidersand their relatives (Araneae Araneoidea Theridiidae) Zool JLinn Soc 141 447ndash626
Agnarsson I Gregoric M Blackledge TA Kuntner M 2012The phylogenetic placement of Psechridae within Entelegynaeand the convergent origin of orb-like spider webs J Zool SystEvol Res 51 100ndash106
Agnarsson I Coddington JA Kuntner M 2013 Systematicsprogress in the study of spider diversity and evolution InPenney D (Ed) Spider Research in the 21st Century Trendsand Perspectives Siri Scientific Press Rochdale UK pp 58ndash111
Alvarez-Padilla F Dimitrov D Giribet G Hormiga G 2009Phylogenetic relationships of the spider family Tetragnathidae(Araneae Araneoidea) based on morphological and DNAsequence data Cladistics 25 109ndash146
Arnedo MA Agnarsson I Gillespie RG 2007 Molecularinsights into the phylogenetic structure of the spider genusTheridion (Araneae Theridiidae) and the origin of the HawaiianTheridion-like fauna Zool Scr 36 337ndash352
Arnedo MA Hormiga G Scharff N 2009 Higher-levelphylogenetics of linyphiid spiders (Araneae Linyphiidae) basedon morphological and molecular evidence Cladistics 25 231ndash262
Ayoub NA Garb JE Hedin M Hayashi CY 2007 Utility ofthe nuclear protein-coding gene elongation factor-1 gamma(EF-1 [c]) for spider systematics emphasizing family levelrelationships of tarantulas and their kin (AraneaeMygalomorphae) Mol Phylogenet Evol 42 394ndash409
Bazinet AL Cummings MP Mitter KT Mitter CW 2013Can RNA-Seq resolve the rapid radiation of advanced mothsand butterflies (Hexapoda Lepidoptera Apoditrysia) Anexploratory study PLoS ONE 8 e82615
Benoit PLG 1962 Les Araneidae-Nephilinae africains RevSuisse Zool 65 217ndash231
Blackledge TA 2012 Spider silk a brief review and prospectus onresearch linking biomechanics and ecology in draglines and orbwebs J Arachnol 40 1ndash12
Blackledge TA Scharff N Coddington JA Szeurouts T WenzelJW Hayashi CY Agnarsson I 2009 Reconstructing webevolution and spider diversification in the molecular era ProcNatl Acad Sci USA 106 5229ndash5234
Bollback JP 2006 SIMMAP Stochastic character mapping ofdiscrete traits on phylogenies BMC Bioinformatics 7 88
Bond JE Garrison NL Hamilton CA Godwin RL HedinM Agnarsson I 2014 Phylogenomics resolves a spider
backbone phylogeny and rejects a prevailing paradigm for orbweb evolution Curr Biol 24 1765ndash1771
Bonnet P 1955 Bibliographia Araneorum Douladoure ToulouseBonnet P 1958 Bibliographia Araneorum 2nd edn Douladoure
ToulouseBouckaert R Heled J Keurouhnert D Vaughan T Wu C-H Xie
D Suchard MA Rambaut A Drummond AJ 2014BEAST 2 A software platform for bayesian evolutionaryanalysis PLoS Comput Biol 10 e1003537
Brignoli PM 1970 Contribution a la connaissance desSymphytognathidae palearctiques (Arachnida Araneae) BullMus Natn His Nat 41 1403ndash1420
Brignoli PM 1983 A Catalogue of the Araneae DescribedBetween 1940 and 1981 Manchester Universtiy PressManchester
Capella-Gutierrez S Silla-Martınez JM Gabaldon T 2009trimAl a tool for automated alignment trimming in large-scalephylogenetic analyses Bioinformatics 25 1972ndash1973
Coddington JA 1986 The monophyletic origin of the orb web InShear WA (Ed) Spiders Webs Behavior and EvolutionStanford University Press Stanford CA USA pp 319ndash363
Coddington JA 1990 Cladistics and spider classificationaraneomorph phylogeny and the monophyly of orbweavers(Araneae Araneomorphae Orbiculariae) Acta Zool Fenn 19075ndash87
Coddington JA Levi HW 1991 Systematics and evolution ofspiders (Araneae) Annu Rev Ecol Syst 22 565ndash592
Dahl F 1906 Die gestreckte Keuroorperform bei Spinnen und dasSystem der Araneen Zool Anz 31 60ndash64
Davies VT 1988 An illustrated guide to the genera of orb-weaving spiders in Australia Mem Queensl Mus 25 273ndash332
Dimitrov D Hormiga G 2009 Revision and cladistic analysis ofthe orbweaving spider genus Cyrtognatha Keyserling 1881(Araneae Tetragnathidae) Bull Am Mus Nat Hist 317 1ndash140
Dimitrov D Hormiga G 2011 An extraordinary new genus ofspiders from Western Australia with an expanded hypothesis onthe phylogeny of Tetragnathidae (Araneae) Zool J Linn Soc161 735ndash768
Dimitrov D Lopardo L Giribet G Arnedo MA Alvarez-Padilla F Hormiga G 2012 Tangled in a sparse spider websingle origin of orb weavers and their spinning work unravelledby denser taxonomic sampling Proc R Soc B 279 1341ndash1350
Dimitrov D Benavides LR Giribet G Arnedo MAGriswold CE Scharff N Hormiga G 2013 Squeezing thelast drops of the lemon molecular phylogeny of Orbiculariae(Araneae) Presented at the 19th International Congress ofArachnology Kenting National Park Taiwan p 47
Dos Reis M Zhu T Yang Z 2014 The impact of the rate prioron bayesian estimation of divergence times with multiple lociSyst Biol 63 555ndash565
Drummond AJ Ho SYW Phillips MJ Rambaut A 2006Relaxed phylogenetics and dating with confidence PLoS Biol 4e88
Eberhard WG 1977 ldquoRectangular orbrdquo webs of Synotaxus(Araneae Theridiidae) J Nat Hist 11 501ndash507
Eberhard WG 1982 Behavioral characters for the higherclassification of orb-weaving spiders Evolution 36 1067ndash1095
Eberhard WG 1995 The web and building behavior of Synotaxusecuadorensis (Araneae Synotaxidae) J Arachnol 23 25ndash30
Exline H Levi HW 1965 The spider genus Synotaxus (AraneaeTheridiidae) Trans Am Microsc Soc 84 177ndash184
Farris JS 1997 The future of phylogeny reconstruction Zool Scr26 303ndash311
Farris JS Albert VA Keuroallersjeuroo M Lipscomb D Kluge AG1996 Parsimony jackknifing outperforms neighbor-joiningCladistics 12 99ndash124
Fernandez R Hormiga G Giribet G 2014 Phylogenomicanalysis of spiders reveals nonmonophyly of orb weavers CurrBiol 24 1772ndash1777
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 247
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
Fitzgerald BM Sirvid PJ 2009 A revision of Nomaua (AraneaeSynotaxidae) and description of a new synotaxid genus fromNew Zealand Tuhinga 20 137ndash158
Forster RR 1955 Spiders of the family Archaeidae from Australiaand New Zealand Trans R Soc NZ 83 391ndash403
Forster RR 1970 The spiders of New Zealand part III DesidaeDictynidae Hahniidae Amaurobioididae Nicodamidae OtagoMus Bull 4 1ndash309
Forster RR 1988 The spiders of New Zealand Part VI FamilyCyatholipidae Otago Mus Bull 6 7ndash34
Forster R Forster L 1999 Spiders of New Zealand and TheirWorldwide kin University of Otago Press Otago New Zealand
Forster RR Platnick NI 1984 A review of the archaeid spidersand their relatives with notes on the limits of the superfamilyPalpimanoidea (Arachnida Araneae) Bull Am Mus Nat Hist178 1ndash106
Forster RR Platnick NI Coddington JA 1990 A proposaland review of the spider family Synotaxidae (AraneaeAraneoidea) with notes on theridid interrelationships Bull AmMus Nat Hist 193 1ndash116
Framenau VW Scharff N Harvey MS 2010 Systematics ofthe Australian orb-weaving spider genus Demadiana withcomments on the generic classification of the Arkyinae (AraneaeAraneidae) Invertebr Syst 24 139ndash171
Gavish-Regev E Hormiga G Scharff N 2013 Pedipalp scleritehomologies and phylogenetic placement of the spider genusStemonyphantes (Linyphiidae Araneae) and its implications forlinyphiid phylogeny Invertebr Syst 27 38ndash52
Goloboff PA 1999 Analyzing large data sets in reasonable timessolutions for composite optima Cladistics 15 415ndash428
Goloboff PA Farris JS Nixon KC 2008 TNT a freeprogram for phylogenetic analysis Cladistics 24 774ndash786
Gregoric M Agnarsson I Blackledge TA Kuntner M 2015Phylogenetic position and composition of Zygiellinae andCaerostris with new insight into orb-web evolution andgigantism Zool J Linn Soc 175 225ndash243
Griswold CE Coddington JA Hormiga G Scharff N 1998Phylogeny of the orb-web building spiders (Araneae OrbiculariaeDeinopoidea Araneoidea) Zool J Linn Soc 123 1ndash99
Griswold CE Coddington JA Platnick NI Forster RR1999 Towards a phylogeny of entelegyne spiders (AraneaeAraneomorphae Entelegynae) J Arachnol 27 53ndash63
Griswold CE Ramırez MJ Coddington JA Platnick NI2005 Atlas of phylogenetic data for entelegyne spiders (AraneaeAraneomorphae Entelegynae) with comments on theirphylogeny Proc Calif Acad Sci 56 1ndash324
Hall TA 1999 BioEdit a user-friendly biological sequencealignment editor and analysis program for Windows 9598NTNucleic Acids Symp Ser 41 95ndash98
Harvey M 1995 The systematics of the spider family Nicodamidae(Araneae Amaurpbioidea) Invertebr Syst 9 279ndash386
Hausdorf B 1999 Molecular phylogeny of araneomorph spiders JEvol Biol 12 980ndash985
Hawthorn AC Opell BD 2003 van der Waals and hygroscopicforces of adhesion generated by spider capture threads J ExpBiol 206 3905ndash3911
Hayashi CY 1996 Molecular Systematics of Spiders Evidencefrom Ribosomal DNA Yale University New Haven CT USA
Heath TA 2012 A hierarchical bayesian model for calibratingestimates of species divergence times Syst Biol 61 793ndash809
Heimer S 1984 Remarks on the spider genus Arcys Walckenaer1837 with description of new species (Araneae Mimetidae)Entomol Abh Dres 47 155ndash178
Heimer S Nentwig W 1983 Thoughts on the phylogeny of theAraneoidea Latreille 1806 (Arachnida Araneae) J Zool SystEvol Res 20 284ndash295
Heimer S Hunter JM Oey TS Levi HW 1982 New sensory() organ on a spider tarsus J Arachnol 10 278ndash279
Hillis DM Pollock DD McGuire JA Zwickl DJ 2003 Issparse taxon sampling a problem for phylogenetic inferenceSyst Biol 52 124ndash126
Hormiga G 2000 Higher level phylogenetics of erigonine spiders(Araneae Linyphiidae Erigoninae) Smithson Contrib Zool609 1ndash160
Hormiga G 2008 On the spider genus Weintrauboa (AraneaePimoidae) with a description of a new species from China andcomments on its phylogenetic relationships Zootaxa 1814 1ndash20
Hormiga G Dimitrov D 2010 Phylogeny of the spider familyPimoidae (Araneoidea) In Book of Abstracts 18th InternationalCongress of Arachnology Presented at the 18th InternationalCongress of Arachnology University of Podlasie amp InternationalSociety of Arachnology Siedlce Poland p 189
Hormiga G Griswold CE 2014 Systematics phylogeny andevolution of orb-weaving spiders Ann Rev Entomol 59 487ndash512
Hormiga G Tu L 2008 On Putaoa a new genus of the spiderfamily Pimoidae (Araneae) from China with a cladistic test of itsmonophyly and phylogenetic placement Zootaxa 1792 1ndash21
Hormiga G Eberhard W Coddington J 1995 Web-constructionbehavior in Australian Phonognatha and the phylogeny ofnepheline and tetragnathid spiders (Araneae Tetragnathidae)Aust J Zool 43 313ndash364
Katoh K Standley DM 2013 MAFFT multiple sequencealignment software version 7 Improvements in performance andusability Mol Biol Evol 30 772ndash780
Katoh K Toh H 2008 Improved accuracy of multiple ncRNAalignment by incorporating structural information into aMAFFT-based framework BMC Bioinform 9 212
Kearney M 2002 Fragmentary taxa missing data and ambiguitymistaken assumptions and conclusions Syst Biol 51 369ndash381
Koch L 1865 Beschreibungen neuer Arachniden und MyriapodenVerh Zool Bot Ges Wien 15 857ndash892
Kuntner M 2006 Phylogenetic systematics of the Gondwanannephilid spider lineage Clitaetrinae (Araneae Nephilidae) ZoolScr 35 19ndash62
Kuntner M Coddington JA Hormiga G 2008 Phylogeny ofextant nephilid orb-weaving spiders (Araneae Nephilidae)testing morphological and ethological homologies Cladistics 24147ndash217
Kuntner M Arnedo MA Trontelj P Lokovsek T AgnarssonI 2013 A molecular phylogeny of nephilid spiders evolutionaryhistory of a model lineage Mol Phylogenet Evol 69 961ndash979
Lartillot N Lepage T Blanquart S 2009 PhyloBayes 3 aBayesian software package for phylogenetic reconstruction andmolecular dating Bioinformatics 25 2286ndash2288
Lehtinen PT 1967 Classification of the cribellate spiders and someallied families with notes on the evolution of the suborderAraneomorpha Ann Zool Fenn 4 199ndash468
Levi HW 1986 The neotropical orb-weaver genera Chrysometaand Homalometa (Araneae Tetragnathidae) El generoneotropical de tejedoras de esferas Chrysometa y Homalometa(Araneae Tetragnathidae) Bull Mus Comp Zool 151 91ndash215
Levi HW Coddington J 1983 Progress report on the phylogenyof the orb-weaving family Araneidae and the superfamilyAraneoidea (Arachnida Araneae) Abh Verh Naturwiss VerHamburg (NF) 26 151ndash154
Levi HW Levi LR 1962 The genera of the spider familyTheridiidae Bull Mus Comp Zool 127 1ndash71
Li SQ Wunderlich J 2008 Sinopimoidae a new spider familyfrom China (Arachnida Araneae) Acta Zootaxon Sin 33 1ndash6
Lopardo L Hormiga G 2008 Phylogenetic placement of theTasmanian spider Acrobleps hygrophilus (Araneae Anapidae)with comments on the evolution of the capture web inAraneoidea Cladistics 24 1ndash33
Lopardo L Hormiga G 2015 Out of the twilight zonephylogeny and evolutionary morphology of the orb-weavingspider family Mysmenidae with a focus on spinneret spigotmorphology in symphytognathoids (Araneae Araneoidea) ZoolJ Linn Soc 173 527ndash786
Lopardo L Giribet G Hormiga G 2011 Morphology to therescue molecular data and the signal of morphological
248 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
characters in combined phylogenetic analysesmdasha case study frommysmenid spiders (Araneae Mysmenidae) with comments on theevolution of web architecture Cladistics 27 278ndash330
Marples RR 1959 The dictynid spiders of New Zealand TransProc R Soc NZ 87 333ndash361
Miller JA Hormiga G 2004 Clade stability and the addition ofdata a case study from erigonine spiders (Araneae LinyphiidaeErigoninae) Cladistics 20 385ndash442
Miller J Griswold C Yin C 2009 The symphytognathoidspiders of the Gaoligongshan Yunnan China (AraneaeAraneoidea) Systematics and diversity of micro-orbweaversZooKeys 11 9ndash195
Miller JA Carmichael A Ramırez MJ Spagna JC HaddadCR Rezac M Johannesen J Kral J Wang X-PGriswold CE 2010 Phylogeny of entelegyne spiders affinitiesof the family Penestomidae (new rank) generic phylogeny ofEresidae and asymmetric rates of change in spinning organevolution (Araneae Araneoidea Entelegynae) Mol PhylogenetEvol 55 786ndash804
Miller MA Pfeiffer W Schwartz T 2011 The CIPRES sciencegateway a community resource for phylogenetic analyses InTGrsquo11 TeraGrid 2011 Salt Lake City UT USA July 18ndash212011 ACM New York NY USA p 41 ISBN 978-1-4503-0888-5 Available at httpdlacmorgcitationcfmid=2016741amppicked=proxampcfid=598348418ampcftoken=93322913
Misof B Liu S Meusemann K Peters RS Donath A MayerC Frandsen PB Ware J Flouri T Beutel RG NiehuisO Petersen M Izquierdo-Carrasco F Wappler T Rust JAberer AJ Aspeuroock U Aspeuroock H Bartel D Blanke ABerger S Beuroohm A Buckley TR Calcott B Chen JFriedrich F Fukui M Fujita M Greve C Grobe P GuS Huang Y Jermiin LS Kawahara AY Krogmann LKubiak M Lanfear R Letsch H Li Y Li Z Li J LuH Machida R Mashimo Y Kapli P McKenna DDMeng G Nakagaki Y Navarrete-Heredia JL Ott M OuY Pass G Podsiadlowski L Pohl H von Reumont BMScheuroutte K Sekiya K Shimizu S Slipinski A StamatakisA Song W Su X Szucsich NU Tan M Tan X TangM Tang J Timelthaler G Tomizuka S Trautwein MTong X Uchifune T Walzl MG Wiegmann BMWilbrandt J Wipfler B Wong TK Wu Q Wu G Xie YYang S Yang Q Yeates DK Yoshizawa K Zhang QZhang R Zhang W Zhang Y Zhao J Zhou C Zhou LZiesmann T Zou S Li Y Xu X Zhang Y Yang HWang J Wang J Kjer KM Zhou X 2014 Phylogenomicsresolves the timing and pattern of insect evolution Science 346763ndash767
Murienne J Edgecombe GD Giribet G 2010 Includingsecondary structure fossils and molecular dating in the centipedetree of life Mol Phylogenet Evol 57 301ndash313
Opell BD 1998 Economics of spider orb-webs the benefits ofproducing adhesive capture thread and of recycling silk FunctEcol 12 613ndash624
Pan H-C Zhou K-Y Song D-X Qiu Y 2004 Phylogeneticplacement of the spider genus Nephila (Araneae Araneoidea)inferred from rRNA and MaSp1 gene sequences Zool Sci 21343ndash351
Paquin P Vink CJ Duperre N 2010 Spiders of New ZealandAnnotated Family key and Species List Manaaki Whenua PressLandcare Research New Zealand
Paradis E 2012 Analysis of Phylogenetics and Evolution with R2nd edn Springer New York
Penney D 2003 Does the fossil record of spiders track that oftheir principal prey the insects Earth Environ Sci Trans RSoc Edinb 94 275ndash281
Penney D 2014 Predatory behaviour of Cretaceous social orb-weaving spiders comment Hist Biol 26 132ndash134
Platnick NI Shadab MU 1993 A review of the pirate spiders(Aranae Mimetidae) of Chile Am Mus Novit 3074 1ndash30
Poinar G Buckley R 2012 Predatory behaviour of the socialorb-weaver spider Geratonephila burmanica n gen n sp
(Araneae Nephilidae) with its wasp prey Cascoscelio incassus ngen n sp (Hymenoptera Platygastridae) in Early CretaceousBurmese amber Hist Biol 24 519ndash525
Pollock DD Zwickl DJ McGuire JA Hillis DM 2002Increased taxon sampling is advantageous for phylogeneticInference Syst Biol 51 664ndash671
Rambaut A Drummond AJ 2007 Tracer v14 Available fromhttpbeastbioedacukTracer accessed on 10 March 2015)
Ramırez MJ 2014 The morphology and phylogeny of dionychanspiders (Araneae Araneomorphae) Bull Am Mus Nat Hist390 1ndash374
Revell LJ 2012 phytools an R package for phylogeneticcomparative biology (and other things) Meth Ecol Evol 3217ndash223
Rix MG 2006 Systematics of the Australasian spider familyPararchaeidae (ArachnidaAraneae) Invertebr Syst 20 203ndash254
Rix M Harvey M 2010a The spider familyMicropholcommatidae (Arachnida Araneae Araneoidea) arelimitation and revision at the generic level ZooKeys 36 1ndash321
Rix MG Harvey MS 2010b The first pararchaeid spider(Araneae Pararchaeidae) from New Caledonia with a discussionon spinneret spigots and egg sac morphology in OzarchaeaZootaxa 2412 27ndash40
Rix MG Harvey MS Roberts JD 2008 Molecularphylogenetics of the spider family Micropholcommatidae(Arachnida Araneae) using nuclear rRNA genes (18S and 28S)Mol Phylogenet Evol 46 1031ndash1048
Roewer CF 1942 Katalog der Araneae von 1758 bis 1940Bremen
Scharff N Coddington JA 1997 A phylogenetic analysis of theorb-weaving spider family Araneidae (Arachnida Araneae)Zool J Linn Soc 120 355ndash434
Scharff N Hormiga G 2012 First evidence of aggressivechemical mimicry in the Malagasy orb weaving spiderExechocentrus lancearius Simon 1889 (Arachnida AraneaeAraneidae) and description of a second species in the genusArthropod Syst Phylogeny 70 107ndash118
Scheuroutt K 2000 The limits of the Araneoidea (Arachnida Araneae) Aust J Zool 48 135ndash153
Scheuroutt K 2003 Phylogeny of Symphytognathidae sl (AraneaeAraneoidea) Zool Scr 32 129ndash151
Selden PA Shih C Ren D 2011 A golden orb-weaver spider(Araneae Nephilidae Nephila) from the Middle Jurassic ofChina Biol Lett 7 775ndash778
Selden PA Shih C Ren D 2013 A giant spider from theJurassic of China reveals greater diversity of the orbicularianstem group Naturwissenschaften 100 1171ndash1181
Simon E 1864 Histoire naturelle des araignees (Araneides)Librairie encyclopedique de Roret Paris
Simon E 1892 Histoire naturelle des araignees 1st edn Librairieencyclopedique de Roret Paris
Simon E 1897 Histoire naturelle des araignees 2nd edn ParisSimon E 1898 Histoire naturelle des araignees ParisSmith SA OrsquoMeara BC 2012 treePL divergence time
estimation using penalized likelihood for large phylogeniesBioinformatics 28 2689ndash2690
Spagna JC Gillespie RG 2006 Unusually long hyptiotes(Araneae Uloboridae) sequence for small subunit (18s)ribosomal rna supports secondary structure model utility inspiders J Arachnol 34 557ndash565
Spagna JC Gillespie RG 2008 More data fewer shiftsmolecular insights into the evolution of the spinning apparatus innon-orb-weaving spiders Mol Phylogenet Evol 46 347ndash368
Spagna JC Crews SC Gillespie RG 2010 Patterns of habitataffinity and AustralHolarctic parallelism in dictynoid spiders(Araneae Entelegynae) Invertebr Syst 24 238ndash257
Stamatakis A 2014 RAxML version 8 a tool for phylogeneticanalysis and post-analysis of large phylogenies Bioinformatics30 1312ndash1313
Vink CJ Duperre N McQuillan BN 2011 The black-headedjumping spider Trite planiceps Simon 1899 (Araneae Salticidae)
Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250 249
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
redescription including cytochrome c oxidase subunit 1 andparalogous 28S sequences NZ J Zool 38 317ndash331
Walckenaer CA 1841 Histoire naturelle des Insects ApteresParis
Wiens JJ 2003 Missing data incomplete taxa and phylogeneticaccuracy Syst Biol 52 528ndash538
Wiens JJ Morrill MC 2011 Missing data in phylogeneticanalysis reconciling results from simulations and empirical dataSyst Biol 60 719ndash731
Wood HM Griswold CE Gillespie RG 2012 Phylogeneticplacement of pelican spiders (Archaeidae Araneae) with insightinto evolution of the ldquoneckrdquo and predatory behaviours of thesuperfamily Palpimanoidea Cladistics 28 598ndash626
Wood HM Matzke NJ Gillespie RG Griswold CE 2013Treating fossils as terminal taxa in divergence time estimationreveals ancient vicariance patterns in the palpimanoid spidersSyst Biol 62 264ndash284
World Spider Catalog (2016) World Spider Catalog NaturalHistory Museum Bern online at httpwscnmbech version170 accessed on 8 January 2016
Wunderlich J 1986 Fossile Spinnen in Bernstein und ihre heutelebenden Verwandten Erich Bauer bei Quelle amp MeyerWiesbaden
Wunderlich J 2004 Fossil spiders (Araneae) of the superfamilyDysderoidea in Baltic and Dominican amber with revised familydiagnoses Beitr Araneol 3 633ndash746
Supporting Information
Additional Supporting Information may be found inthe online version of this articleFigure S1 Maximum likelihood of a version of the
dataset that includes 28S and 18S sequences that mightbe problematic (see supporting information for discus-sion)Figure S2 Results from the Bayesian analyses of the
full dataset (in PhyloBayes)Figure S3 Full topology from the maximum likeli-
hood analyses of the full dataset (simplified version ispresented in Fig 2)Figure S4 Full topology from the maximum likeli-
hood analyses of the matrix_3g dataset
Figure S5 Full topology from the maximum likeli-hood analyses of the matrix_4g datasetFigure S6 Full topology from the maximum likeli-
hood analyses of the dataset treated with trimAL toremove ambiguously aligned regionsFigure S7 Single best topology from the maximum
parsimony analyses of the full dataset(length = 298012)Figure S8 Results from the molecular dating analy-
ses in BEAST using constraint of Araneidae excludingNephilinaeFigure S9 Results from the molecular dating analy-
ses in BEAST using constraint of the redefined Aranei-dae (including Nephilinae) and an unpartitioneddatasetFigure S10 Web architecture evolutionary historymdash
summary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingFigure S11 Cribellum evolutionary historymdashsum-
mary of 1000 SIMMAP characters maps using thedated tree based on Araneidae excluding NephilinaedatingTable S1 GenBank accession numbers for all
sequences used in this studyTable S2 List of the different data matrices with the
corresponding number of taxa and charactersTable S3 List of fossil constraints and the relevant
settings for the fossils constantans implementation inthe molecular dating analysesTable S4 Summary of the support under different
analytical treatments for the main clades discusses inthe present manuscriptData S1 traitsnexmdashannotated nexus with the infor-
mation for the web and cribellum used in the compar-ative analyses
250 Dimitar Dimitrov et al Cladistics 33 (2017) 221ndash250
