BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors nonprofit publishers academic institutions researchlibraries and research funders in the common goal of maximizing access to critical research

EXCEPTIONALLY PRESERVED SKELETONS OF THE CRETACEOUSSNAKE DINILYSIA PATAGONICA WOODWARD 1901Author(s) MICHAEL W CALDWELL and ADRIANA ALBINOSource Journal of Vertebrate Paleontology 22(4)861-866 2002Published By The Society of Vertebrate PaleontologyDOI httpdxdoiorg1016710272-4634(2002)022[0861EPSOTC]20CO2URL httpwwwbiooneorgdoifull1016710272-4634282002290225B08613AEPSOTC5D20CO3B2

BioOne (wwwbiooneorg) is a nonprofit online aggregation of core research in the biological ecological andenvironmental sciences BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies associations museums institutions and presses

Your use of this PDF the BioOne Web site and all posted and associated content indicates your acceptance ofBioOnersquos Terms of Use available at wwwbiooneorgpageterms_of_use

Usage of BioOne content is strictly limited to personal educational and non-commercial use Commercial inquiriesor rights and permissions requests should be directed to the individual publisher as copyright holder

861

Journal of Vertebrate Paleontology 22(4)861ndash866 December 2002q 2002 by the Society of Vertebrate Paleontology

RAPID COMMUNICATTION

EXCEPTIONALLY PRESERVED SKELETONS OF THE CRETACEOUS SNAKEDINILYSIA PATAGONICA WOODWARD 1901

MICHAEL W CALDWELL1 and ADRIANA ALBINO2

1Department of Earth and Atmospheric Sciences and Department of Biological Sciences University of Alberta EdmontonAlberta T6G 2E3 Canada mwcaldwellualbertaca

2Consejo Nacional de Investigaciones Cientıficas y Tecnicas (CONICET) Departamento de Biologıa Universidad Nacional deMar del Plata Funes 3250 7600 Mar del Plata Argentina

For over one hundred years the Upper Cretaceous snakeDinilysia patagonica has been known from only the holotypeskull and associated vertebrae (Smith-Woodward 1901 Esteset al 1970 Frazetta 1970 Hecht 1982 Rage and Albino1989 Albino 1996) Recent fieldwork near Neuquen Argen-tina has produced a large number of exceptionally well-pre-served skulls and skeletons of Dinilysia that preserve osteo-logical features missing in the holotype This new informationis timely as recent studies of snake phylogeny have reliedupon incomplete descriptions of Dinilysia (Caldwell 1999Scanlon and Lee 2000 Tchernov et al 2000) Our study ofthe new fossils indicates that Dinilysia was a large-bodiedsnake with an aniliid xenopeltid-like middle-ear osteologyand an anguimorph-like hypapophysealintercentrum anatomyThese new data will certainly impact future studies of snakephylogeny We provide the first diagnosis of Dinilysia pata-gonica Smith-Woodward 1901 along with a preliminary de-scription of the new material followed by discussion of twoimportant osteological features (ie postorbital vs jugal ab-sence of a crista circumfenestralis)

The holotype and all new specimens of Dinilysia describedhere were found at three localities in or near Neuquen Ne-uquen Province Argentina The fossil-producing units arered- to white-weathering coarse-grained sandstones assignedto the Bajo de la Carpa Member Rio Colorado Formation(Coniacian Upper Cretaceous Caldwell and Albino 2001)The holotype specimen (Smith-Woodward 1901 Estes et al1970) and two fragmentary skulls were found at Boca delSapo just north of Neuquen Another fragmentary skull as-signed here to Dinilysia was collected in the 1980s fromoutcrops just north of the Universidad Nacional del Comahuein Neuquen The most recent collection of excellent skullsand articulated postcranial skeletons is from outcrops of theRıo Colorado Formation exposed at the lsquolsquoTripailao Farm Lo-calityrsquorsquo These sections are located on the south side of theRio Negro approximately 85 km to the west of the bridge atPaso Cordoba or approximately 23 km to the southeast ofNeuquen on the south side of the river (Caldwell and Albino2001)

Institutional Abbreviations MACN Museo Argentino deCiencias Naturales lsquolsquoBernardino Rivadaviarsquorsquo Buenos Aires Ar-gentina MLP Museo de La Plata La Plata ArgentinaMPCA-PV Museo de la Ciudad de Cipoletti Cipoletti RıoNegro Province Argentina MUCP Museo de Geologıa y Pa-leontologıa Universidad Nacional del Comahue Neuquen Ar-gentina

SYSTEMATIC PALEONTOLOGY

REPTILIA Linnaeus 1758SQUAMATA Oppel 1811

SERPENTES Linnaeus 1758DINILYSIIDAE McDowell 1987

DINILYSIA Smith-Woodward 1901DINILYSIA PATAGONICA Smith-Woodward 1901

(Figs 1ndash4)

Holotype MLP 26-410 a partial skull (Figs 1A 3B C)and partial postcranium

Locality and Horizon Locality at Boca del Sapo Neu-quen Argentina (Coniacian Upper Cretaceous)

Diagnosis Medium-sized snake uniquely possessing com-plex interdigitating frontal-parietal suture maxilla with deepanterolaterally directed trough on suborbital surface palatineand prefrontal forming broad ventral facet for articulation withmaxilla and also forming choanal groove ventromedial processof coronoid contacting angular triradiate postfrontal overlap-ping frontalndashparietal suture postorbital closing posterior orbitalmargin with distinct lsquofootrsquo articulating with maxillary troughlarge lsquorsquo-shaped quadrate stapes robust with extremely largeand expanded stapedial footplate extracolumellaintercalary el-ement contacting quadrate suprastapedial process unfused in-tercentra present on hypapophyses of thirdndashfourth precloacalsfused hypapophysesintercentra to at least 10th precloacal an-terior 13 of precloacals bearing prominent ventral hypapophy-seal lsquolsquokeelsrsquorsquo

Referred Specimens MACN-RN 1013 (isolated skullFigs 1B 3A) MACN-RN 1014 (isolated skull Figs 1C 3DndashG) MACN-RN 976 (articulated partial skeleton Fig 2A withvery complete skull Fig 1D) MACN-RN 1015 (isolated skullFig 1E) MACN-RN 1016 (29 vertebrae possibly caudals)MACN-RN 1017 (45 dorsal vertebrae) MACN-RN 1018 (eightlarge dorsal vertebrae) and MACN-RN 1021 (35 dorsal ver-tebrae) MPCA-PV 527 (single skull with associated partial po-stcranium Figs 1F 4) from Tripailao Farm Locality Paso Cor-doba Rıo Negro Province Argentina (Coniacian Upper Cre-taceous) MUCP v 38 (fragmentary skull Fig 1G) MACN-N26 (isolated strings of vertebrae associated with MUCP v 38)MUCP v 39 40 98ndash102 104ndash116 119 121 (isolated vertebraeand short articulated strings) north of the Universidad Nacion-al del Comahue Neuquen Argentina (Coniacian Upper Cre-taceous) MLP 79-II-27-1 (fragmentary skull Fig 1H) MLP71-VII-29-1 (fragmentary skull Fig 1I) MLP 79-11-27-2 to79-11-27-17 (articulated strings of vertebrae) from type local-ity at Boca del Sapo Neuquen Argentina (Coniacian UpperCretaceous)

862 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

FIGURE 1 Skulls of Dinilysia patagonica A dorsal view of holotype skull MLP 26-410 Scale equals 1 cm B dorsal view of skull of MACN-RN-1013 inset box with line drawings of MACN-RN-1013 in dorsal ventral and right lateral views and lateral and medial views of rightmandible C dorsal view of skull of MACN-RN-1014 D dorsal view of skull of MACN-RN-976 E dorsal view of skull of MACN-RN-1015F dorsal view of skull of MPCA-PV 527 G dorsal view of skull of MUCP v 38 H dorsal view of eroded skull of MLP 79-II-27-1 I dorsalview of eroded skull of MLP 71-VII-29-1 AndashG Scale equals 1 cm H I Scale equals 5 mm Abbreviations BsOC basioccipital cen centrumns neural spine Co coronoid CpB compound bone D dentary EcP ectopterygoid ex-col extracolumella Fr frontal fld foramen for thelachrymal duct fPal-Mx Palatine-Maxillary foramen fRO fenestra rotundum Mx maxilla Na nasal Pal palatine Par parietal PrF prefrontaPo postorbital Pf postfrontal Pr prootic Pt pterygoid Q quadrate SOC supraoccipital BsP basisphenoid St supratemporal St-F stapedialfootplate St-P stapedial shaftprocess VI 6th cranial nerve (abducens) X 10th cranial nerve (vagus)

DESCRIPTION

Cranial The new Dinilysia material includes osteologicalfeatures that were either not preserved or have since been lostin the holotype specimen (Smith-Woodward 1901 Estes et al1970) and we have chosen to focus our description on thesenew features Some elements such as the premaxilla remainunknown (Fig 1B)

The maxilla is long and low in lateral view has at leastfourteen tooth positions and three maxillary foramina and isrounded on its anterior tip At the midpoint of the maxilla thereis a prominent medial process that articulates along a slopingfacet with the prefrontal anteriorly and palatine posteriorly (Fig2AndashC) Posterior to the palatineprefrontal process the maxillabears a broad groove for articulation with the foot of the post-orbital Posterior to the postorbital foot the maxilla expands and

863CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 2 Details of skull features of Dinilysia patagonica A right lateral view of skull of MACN-RN 1013 note loose contact of maxillawith ectopterygoid postorbital and palatineprefrontal B line drawing in dorsal view of holotype (MLP 26ndash410) left maxilla postorbital andpalatineprefrontal C line drawing in ventral view of holotype (MLP 26ndash410) left maxilla and palatineprefrontal D left posteroventral viewof skull (MACN-RN 1014) emphasizing size of stapedial footplate and absence of a crista circumfenestralis E shaded line drawing of braincasein left posteroventral view Scale equals 1 cm FndashG Right ventrolateral view and line drawing of skull of MACN-RN 1014 Note preservationand articulations of stapedial shaftprocess (with suprastapedial process of enlarged quadrate) Abbreviations see Figure 1

flattens where it underlies the short robust ectopterygoid in abroad non-sutural articulation the ectopterygoid is broad andsmooth at its contact with the maxilla

The prefrontal is a small pyramid-shaped element that isbroadest along its contact with the maxilla and narrowest at itsarticulation with the frontal where the apex inserts into a shal-low v-shaped facet There is no dorsal and anterior process ofthe prefrontal (Fig 1B) Uniquely the prefrontal contributes tothe palate Along its ventral surface just anterior to the contactwith the palatine the prefrontal forms a channel that is contin-

uous with the choanal channel on the palatine (Fig 2C) andappears to be unique within squamates

The postfrontal is a small triangular bone that articulates withthe frontal and parietal and clasps the tightly sutured fronto-parietal articulation (Fig 1A B D G) the postfrontal entersthe superior margin of the orbit but does not contact the pre-frontal The postorbital is dorsoventrally elongate and shapedlike an inverted lsquolsquoTrsquorsquo The dorsal tip loosely contacts the pos-terior margin of the postfrontal and sidewall of the parietal (Fig2A) while the ventral tip broadened into a footplate-like struc-

864 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ture sits in the dorsal maxillary groove (Figs 1B 2A) Thelsquolsquojugalrsquorsquo that Estes et al (1970) identified in Dinilysia is theventral tip of the postorbital Neither a jugal nor a lachrymalcan be identified in any of the known specimens of Dinilysia

The stapedial footplate is extremely large completely ob-scures the fenestra ovalis (fenestra vestibularis) and is separat-ed from the fenestra rotundum by a low ridge of bone derivedfrom the exoccipital the crista interfenestralis (Fig 2D E) Thecolumellar shaft is at an acute angle relative to the footplateand is directed posteriorly towards the suprastapedial processof the quadrate It does not articulate directly with the supras-tapedial process but instead with a round ossified element weidentify as the extracolumella (Fig 2F G) this element has alsobeen termed the intercalary or stylohyal (for recent use of theseterms see Frazetta 1999 or Rieppel 1980a) The quadrate is alarge lsquoquestion markrsquo-shaped element with a broad cephaliccondyle robust shaft and broad mandibular condyle The su-prastapedial process is very large and expanded mediolaterallyas well as ventrally Articulation of the extracolumella-colu-mella complex is with the extreme ventromedial tip of the su-prastapedial process

Postcranial Two of the new postcranial skeletons are par-ticularly informative (Fig 3AndashE) The more complete and ar-ticulated specimen MACN RN 976 preserves 123 vertebrae ina series of fractured but associated blocks (Fig 3A) The lon-gest complete string of 40 vertebrae consists of anteriormostprecloacals that are still articulated with the skull (see Rage andAlbino 1989 for a description of the mid-precloacal vertebrae)All the vertebrae bear paracotylar foramina Unsuspected ver-tebral features are found on the newest Dinilysia specimen(MPCAndashPV 527) The third and fourth most anterior pre-cloa-cals bear large posteriorly positioned hypapophyses with largeunfused intercentra (Fig 3B) The intercentrum of the fourthprecloacal has been lost thus revealing the deeply concave sur-face of the hypapophysis (Fig 3B) The anterior hypapophysesof Dinilysia are remarkable as no known fossil or extant snakepossesses hypapophyses with unfused intercentra The posteriorposition of the hypapophyses and presence of unfused intercen-tra are also observed in mosasaurs (Russell 1967) dolichosaurs(Caldwell 2000) and varanoid lizards (Rieppel 1980b) Incontrast the fifth most-anterior precloacal bears a large hypa-pophysis likely with a fused intercentrum that is typical ofsnakes (Fig 3B)

DISCUSSION

Postfrontal and Postorbital We identify a postfrontal andpostorbital in Dinilysia While a postorbital is common to mostmodern snakes a separate postfrontal is not Estes et al (1970)identified the latter bone in Dinilysia as homologous to the post-frontal in lizards We concur with this identification as the post-frontal of Dinilysia passes the positional and relational criteriaof the test of similarity (ie element clasping the fronto-parietalsuture) and the test of composition (ie dermatocranial bone)Our identification of a postorbital in Dinilysia follows the con-vention for snakes of identifying the element that descends ven-trally from the area of the frontal-parietal suture and framingthe posterior margin of the orbit as the postorbital Though wefollow this convention for Dinilysia we do find this identifi-cation problematic due to the presence of the postfrontal andthe jugal-like articulations of the postorbital of Dinilysia

In non-snake squamates the postorbital contacts the postfron-tal sometimes fuses with it extends a posterior ramus to con-tact the squamosal and contacts the jugal where that elementrises to meet the posterior margin of the postfrontal In contrastthe jugal bone variably articulates with the postfrontal postor-bital or postorbital frontal articulates with the maxilla behindand beneath the orbit and articulates with the lachrymal if pre-

sent The jugal bone of the varanoid lizard Lanthanotus artic-ulates with the postfrontal maxilla-ectopterygoid and lachry-mal (the postorbital is absent in Lanthanotus) The remainingarticulation ie with the ectopterygoid is variable among liz-ards and snakes (eg in Python the lsquolsquopostorbitalrsquorsquo contacts theectopterygoid) Comparisons to snakes such as scolecophidiansare not possible as all superior and posterior orbital and tem-poral-arch bones are absent The test of topological relationsuggests a closer lsquofitrsquo with the jugal of non-snake squamatesnot the postorbital The possible presence of a jugal in Dinilysiais relevant to recent arguments on the presence or absence ofa jugal in Pachyrhachis (Caldwell and Lee 1997 Zaher andRieppel 1999) and other legged snakes such as Haasiophis(Tchernov et al 2000) Zaher and Rieppel (1999) have claimedthat their studies have lsquolsquorevealedrsquorsquo the jugal of Pachyrhachis tobe a broken ectopterygoid Tchernov et al (2001) using CTScan reconstructions argue that the bone is an enlarged post-orbital that is symmetrically fractured on both sides of the ho-lotype skull While it seems unlikely that the element in ques-tion is an ectopterygoid (see the ectopterygoid of the paratypeskull illustrated by Lee and Caldwell 1998) it is intriguing thatTchernov et al (2001) would identify an element with devel-oped jugal-like topological connectivities as a postorbital sim-ilar to the condition we note in Dinilysia

Stapes and Crista Circumfenestralis As was noted by Es-tes et al (1970) Dinilysia does not possess an identifiable cristacircumfenestralis contra Rage (1984) Rieppel (1988) andTchernov et al (2000) nor is the crista prootica projecting lat-erally around the fenestra ovalis (contra Rieppel and Zaher2001) The enormous size of the stapedial footplate (Fig 3DE) may account for the absence of all cristae ie the cristacircumfenestralis if ad hoc morphogenetic explanations are in-voked (as has been argued for Acrochordus by Rieppel andZaher 2001) It may also be that Dinilysia is plesiomorphicallylacking the crista circumfenestralis as was argued by Estes etal (1970) and McDowell (1974) However at this point it canonly be stated prior to the test of congruence that the test ofsimilarity indicates that the crista circumfenestralis is absent inDinilysia

Unlike the condition in snakes (see Wever 1978) we alsofollow Estes et al (1970) by identifying the opening ventral tothe fenestra ovalis as the fenestra rotundum (Fig 2) The fe-nestra rotundum is present in most lizards but is consideredabsent in snakes and other squamates and turtles that possess areentrant fluid circuit the circuit cannot function if the roundwindow membrane is present (Wever 1978) A foramen orniche (see Wever 1978) is still present as in almost all squa-mates but in snakes it forms the channel of fluid movementfrom the recessus scala tympani to the pericapsular space Inthe otic region of Dinilysia Estes et al (1970) identified thefenestra ovalis the fenestra rotundum and the foramen for the10th cranial nerve (Fig 2D E) The difficulty with identificationof Estes et alrsquos fenestra rotundum as the lateral aperture of therecessus scala tympani is that it is not within the crista circum-fenestralis (the usual condition for extant snakes) In some basalsnakes where the crista circumfenestralis is incomplete poster-oventrally the aperture is referred to as the fenestra pseudoro-tundum

The osteology of the middle ear is also very intriguing Com-parison of the new specimens to the modern macrostomatanXenopeltis (see Frazetta 1999) indicates a number of sharedcharacters Both Dinilysia and Xenopeltis possess a broad sta-pedial footplate with a posterodorsally angled stapedial shaftThat shaft contacts a small extracolumellaintercalarystylohyalelement that articulates with the suprastapedial process of thequadrate The head of the quadrate in both taxa is very broadand laterally flattened along the contact with the supratemporalThe topographic connectivities of the supratemporal are com-

865CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 3 Postcranial skeletal elements of Dinilysia patagonica A dorsal view of articulated specimen of Dinilysia patagonica (MACN-RN-976) scale equals 10 cm B ventral view of the lsquolsquocervicalrsquorsquo series (anteriormost precloacals) of Dinilysia patagonica (MPCA-PV 527) scaleequals 1 cm CndashE anterior dorsal vertebra from holotype in lateral anterior and posterior view (MLP 26ndash410) scale equals 1 cm Abbreviationscen centrum con condyle cot cotyle d-ap diapophyses hk haemal keel na neural arch ns neural spine pactf paracotylar foramen p-apparapophyses poz postzygaophyses prz prezygapophyes przap prezygapophyseal accessory process ratna right atlas neural arch sfsubcentral foramina sg subcentral groove zgf zygantral facets zpf zygosphenal facets

mon in both snakes (ie articulating with the parietal prooticand exoccipital in a broad lsquolsquoLrsquorsquo-shaped groove) Modern aniliidsalso possess a similar osteology though the ossified intercalaryextracolumellar element is absent in Anilius (Rieppel 1980b)

Reference to the phylogenies present in the literature wouldsuggest that the absence of a crista circumfenestralis in Dini-lysia is plesiomorphic However recent descriptions of the oticregion of Achrochordus (Rieppel and Zaher 2001) would sug-gest that absence of the feature in some snake taxa may beexplained by paedomorphosis

The many new features of Dinilysia such as the presence ofa stylohyal element the absence of a crista circumfenestralisor the presence of unfused cervical intercentra raise new ques-tions about the phylogenetic relationships of snakes Questionson primitive versus derived characters for snakes center on thephylogenetic position of scolecophidians the Cretaceous ma-rine snake taxa such as Pachyrhachis and Haasiophis and earlyterrestrial Gondwanan snakes such as Dinilysia and the Madt-soiidae (Caldwell and Lee 1997 Rage and Escuille 2000Scanlon and Lee 2000 Tchernov et al 2000)

866 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ACKNOWLEDGMENTS

We thank P Posse L Salgado J Calvo and S Heredia forassistance while working and living in Neuquen We are in-debted to J Bonaparte for access to collections at the MACNand to R Pascual for permission to study Dinilysia specimensin the MLP We thank J Rose and L A Lindoe for skilledpreparation of the specimens We thank G Hanke R HolmesH Larsson A Nicholson R Nydam and S Pierce We ac-knowledge financial support from a CMN RAC Grant to MCNSERC Grant 238458-01 to MC CONICET for support of AAand a National Geographic CRE Grant for fieldwork support

LITERATURE CITED

Albino A 1996 The South American fossil Squamata (Reptilia Lep-idosauria) pp 9ndash 72 in G Arratia (ed) Contributions of SouthernSouth America to Vertebrate Paleontology Munchner Geowissenschaften Abhandlungen (A) 30

Caldwell M W 1999 Squamate phylogeny and the relationships ofsnakes and mosasauroids Zoological Journal of the Linnean So-ciety 125115ndash147

mdashmdashmdash 2000 An aquatic squamate reptile from the English ChalkDolichosaurus longicollis Owen 1850 Journal of Vertebrate Pa-leontology 20720ndash735

mdashmdashmdash and A Albino 2001 Palaeoenvironment and palaeocology ofthree Cretaceous snakes Pachyophis Pachyrhachis and DinilysiaActa Palaeontologica Polonica 46203ndash218

mdashmdashmdash and M S Y Lee 1997 A snake with legs from the marineCretaceous of the Middle East Nature 386705ndash709

Estes R T H Frazzetta and E E Williams 1970 Studies on thefossil snake Dinilysia patagonica Woodward Part 1 Cranial mor-phology Bulletin of the Museum of Comparative Zoology 14025ndash74

Frazzetta T H 1970 Studies on the fossil snake Dinilysia patagonicaWoodward Part II Jaw machinery in the earliest snakes Forma etFunctio 3205ndash221

mdashmdashmdash 1999 Adaptations and significance of the cranial feeding ap-paratus of the Sunbeam snake (Xenopeltis unicolor) Part 1 Anat-omy of the skull Journal of Morphology 23927ndash43

Hecht M 1982 The vertebral morphology of the Cretaceous snakeDinilysia patagonica Woodward Neues Jarhbuch fur Geologie undPalaontologie Monatschefte 1982523ndash532

Lee M S Y and M W Caldwell 1998 Anatomy and relationshipsof Pachyrhachis a primitive snake with hindlimbs PhilosophicalTransactions of the Royal Society Series B 3531521ndash1552

McDowell S B 1974 A catalogue of the snakes of New Guinea andthe Solomons with special reference to those in the Bernice P

Bishop museum Part 1 Scolecophidia Journal of Herpetology 81ndash57

mdashmdashmdash 1987 Systematics pp 3ndash50 in R A Siegel J T Collins andS S Novak (eds) Snakes Ecology and Evolutionary BiologyMacmillan New York

Rage J-C 1984 Serpentes Handbuch der Palaoherpetologie Teil 11Stuttgart Gustav Fischer Verlag 80 pp

mdashmdashmdash and A Albino 1989 Dinilysia patagonica (Reptilia Serpen-tes) materiel vertebral additionnel du Cretace superieurdrsquoArgentine Etude complementaire des vertebres variations in-traspecifiques et intracolumnaires Neues Jarhbuch fur Geologieund Palaontologie Monatschefte 1989433ndash447

mdashmdashmdash and F Escuillie 2000 Un nouveau serpent bipede du Ceno-manien (Cretace) Implications phyletiques Comptes Rendus delrsquoAcademie des Sciences Paris Serie Ia 3301ndash8

Rieppel O 1980a The Phylogeny of Anguinomorph Lizards BaselBirkhauser Verlag 86 pp

mdashmdashmdash 1980b The sound-transmitting apparatus in primitive snakesand its phylogenetic significance Zoomorphology 9645ndash62

mdashmdashmdash 1988 A review of the origin of snakes Evolutionary Biology2237ndash130

mdashmdashmdash and H Zaher 2001 The development of the skull in Acro-chordus granulatus (Schneider) (Reptilia Serpentes) with specialconsideration of the oticondashoccipital complex Journal of Morphol-ogy 249252ndash266

Russell D 1967 Systematics and morphology of American mosasaursPeabody Museum of Natural History Yale University Bulletin 231ndash241

Scanlon J D and M S Y Lee 2000 The Pleistocene serpent Won-ambi and the early evolution of snakes Nature 403416ndash420

Smith-Woodward A 1901 On some extinct reptiles from Patagonia ofthe genera Miolania Dinilysia and Genyodectes Proceedings ofthe Zoological Society London 1901169ndash184

Tchernov E M J Polcyn and L L Jacobs 2001 Application ofcomputed tomography to Pachyrhachis problematicus a snakewith legs from the Cretaceous of Israel Journal of Vertebrate Pa-leontology 21(3 suppl)107A

mdashmdashmdash O Rieppel H Zaher M J Polcyn and L L Jacobs 2000 Afossil snake with limbs Science 2872010ndash2012

Wever E G 1978 The Reptile Ear Its Structure and Function Prince-ton University Press Princeton 1024 pp

Zaher H and O Rieppel 1999 The phylogenetic relationships of Pa-chyrhachis problematicus and the evolution of limblessness insnakes (Lepidosauria Squamata) Comptes Rendus de lrsquoAcademiedes Sciences Paris 329831ndash837

Received 21 March 2002 accepted 17 June 2002

862 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

FIGURE 1 Skulls of Dinilysia patagonica A dorsal view of holotype skull MLP 26-410 Scale equals 1 cm B dorsal view of skull of MACN-RN-1013 inset box with line drawings of MACN-RN-1013 in dorsal ventral and right lateral views and lateral and medial views of rightmandible C dorsal view of skull of MACN-RN-1014 D dorsal view of skull of MACN-RN-976 E dorsal view of skull of MACN-RN-1015F dorsal view of skull of MPCA-PV 527 G dorsal view of skull of MUCP v 38 H dorsal view of eroded skull of MLP 79-II-27-1 I dorsalview of eroded skull of MLP 71-VII-29-1 AndashG Scale equals 1 cm H I Scale equals 5 mm Abbreviations BsOC basioccipital cen centrumns neural spine Co coronoid CpB compound bone D dentary EcP ectopterygoid ex-col extracolumella Fr frontal fld foramen for thelachrymal duct fPal-Mx Palatine-Maxillary foramen fRO fenestra rotundum Mx maxilla Na nasal Pal palatine Par parietal PrF prefrontaPo postorbital Pf postfrontal Pr prootic Pt pterygoid Q quadrate SOC supraoccipital BsP basisphenoid St supratemporal St-F stapedialfootplate St-P stapedial shaftprocess VI 6th cranial nerve (abducens) X 10th cranial nerve (vagus)

DESCRIPTION

Cranial The new Dinilysia material includes osteologicalfeatures that were either not preserved or have since been lostin the holotype specimen (Smith-Woodward 1901 Estes et al1970) and we have chosen to focus our description on thesenew features Some elements such as the premaxilla remainunknown (Fig 1B)

The maxilla is long and low in lateral view has at leastfourteen tooth positions and three maxillary foramina and isrounded on its anterior tip At the midpoint of the maxilla thereis a prominent medial process that articulates along a slopingfacet with the prefrontal anteriorly and palatine posteriorly (Fig2AndashC) Posterior to the palatineprefrontal process the maxillabears a broad groove for articulation with the foot of the post-orbital Posterior to the postorbital foot the maxilla expands and

863CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 2 Details of skull features of Dinilysia patagonica A right lateral view of skull of MACN-RN 1013 note loose contact of maxillawith ectopterygoid postorbital and palatineprefrontal B line drawing in dorsal view of holotype (MLP 26ndash410) left maxilla postorbital andpalatineprefrontal C line drawing in ventral view of holotype (MLP 26ndash410) left maxilla and palatineprefrontal D left posteroventral viewof skull (MACN-RN 1014) emphasizing size of stapedial footplate and absence of a crista circumfenestralis E shaded line drawing of braincasein left posteroventral view Scale equals 1 cm FndashG Right ventrolateral view and line drawing of skull of MACN-RN 1014 Note preservationand articulations of stapedial shaftprocess (with suprastapedial process of enlarged quadrate) Abbreviations see Figure 1

flattens where it underlies the short robust ectopterygoid in abroad non-sutural articulation the ectopterygoid is broad andsmooth at its contact with the maxilla

The prefrontal is a small pyramid-shaped element that isbroadest along its contact with the maxilla and narrowest at itsarticulation with the frontal where the apex inserts into a shal-low v-shaped facet There is no dorsal and anterior process ofthe prefrontal (Fig 1B) Uniquely the prefrontal contributes tothe palate Along its ventral surface just anterior to the contactwith the palatine the prefrontal forms a channel that is contin-

uous with the choanal channel on the palatine (Fig 2C) andappears to be unique within squamates

The postfrontal is a small triangular bone that articulates withthe frontal and parietal and clasps the tightly sutured fronto-parietal articulation (Fig 1A B D G) the postfrontal entersthe superior margin of the orbit but does not contact the pre-frontal The postorbital is dorsoventrally elongate and shapedlike an inverted lsquolsquoTrsquorsquo The dorsal tip loosely contacts the pos-terior margin of the postfrontal and sidewall of the parietal (Fig2A) while the ventral tip broadened into a footplate-like struc-

864 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ture sits in the dorsal maxillary groove (Figs 1B 2A) Thelsquolsquojugalrsquorsquo that Estes et al (1970) identified in Dinilysia is theventral tip of the postorbital Neither a jugal nor a lachrymalcan be identified in any of the known specimens of Dinilysia

The stapedial footplate is extremely large completely ob-scures the fenestra ovalis (fenestra vestibularis) and is separat-ed from the fenestra rotundum by a low ridge of bone derivedfrom the exoccipital the crista interfenestralis (Fig 2D E) Thecolumellar shaft is at an acute angle relative to the footplateand is directed posteriorly towards the suprastapedial processof the quadrate It does not articulate directly with the supras-tapedial process but instead with a round ossified element weidentify as the extracolumella (Fig 2F G) this element has alsobeen termed the intercalary or stylohyal (for recent use of theseterms see Frazetta 1999 or Rieppel 1980a) The quadrate is alarge lsquoquestion markrsquo-shaped element with a broad cephaliccondyle robust shaft and broad mandibular condyle The su-prastapedial process is very large and expanded mediolaterallyas well as ventrally Articulation of the extracolumella-colu-mella complex is with the extreme ventromedial tip of the su-prastapedial process

Postcranial Two of the new postcranial skeletons are par-ticularly informative (Fig 3AndashE) The more complete and ar-ticulated specimen MACN RN 976 preserves 123 vertebrae ina series of fractured but associated blocks (Fig 3A) The lon-gest complete string of 40 vertebrae consists of anteriormostprecloacals that are still articulated with the skull (see Rage andAlbino 1989 for a description of the mid-precloacal vertebrae)All the vertebrae bear paracotylar foramina Unsuspected ver-tebral features are found on the newest Dinilysia specimen(MPCAndashPV 527) The third and fourth most anterior pre-cloa-cals bear large posteriorly positioned hypapophyses with largeunfused intercentra (Fig 3B) The intercentrum of the fourthprecloacal has been lost thus revealing the deeply concave sur-face of the hypapophysis (Fig 3B) The anterior hypapophysesof Dinilysia are remarkable as no known fossil or extant snakepossesses hypapophyses with unfused intercentra The posteriorposition of the hypapophyses and presence of unfused intercen-tra are also observed in mosasaurs (Russell 1967) dolichosaurs(Caldwell 2000) and varanoid lizards (Rieppel 1980b) Incontrast the fifth most-anterior precloacal bears a large hypa-pophysis likely with a fused intercentrum that is typical ofsnakes (Fig 3B)

DISCUSSION

Postfrontal and Postorbital We identify a postfrontal andpostorbital in Dinilysia While a postorbital is common to mostmodern snakes a separate postfrontal is not Estes et al (1970)identified the latter bone in Dinilysia as homologous to the post-frontal in lizards We concur with this identification as the post-frontal of Dinilysia passes the positional and relational criteriaof the test of similarity (ie element clasping the fronto-parietalsuture) and the test of composition (ie dermatocranial bone)Our identification of a postorbital in Dinilysia follows the con-vention for snakes of identifying the element that descends ven-trally from the area of the frontal-parietal suture and framingthe posterior margin of the orbit as the postorbital Though wefollow this convention for Dinilysia we do find this identifi-cation problematic due to the presence of the postfrontal andthe jugal-like articulations of the postorbital of Dinilysia

In non-snake squamates the postorbital contacts the postfron-tal sometimes fuses with it extends a posterior ramus to con-tact the squamosal and contacts the jugal where that elementrises to meet the posterior margin of the postfrontal In contrastthe jugal bone variably articulates with the postfrontal postor-bital or postorbital frontal articulates with the maxilla behindand beneath the orbit and articulates with the lachrymal if pre-

sent The jugal bone of the varanoid lizard Lanthanotus artic-ulates with the postfrontal maxilla-ectopterygoid and lachry-mal (the postorbital is absent in Lanthanotus) The remainingarticulation ie with the ectopterygoid is variable among liz-ards and snakes (eg in Python the lsquolsquopostorbitalrsquorsquo contacts theectopterygoid) Comparisons to snakes such as scolecophidiansare not possible as all superior and posterior orbital and tem-poral-arch bones are absent The test of topological relationsuggests a closer lsquofitrsquo with the jugal of non-snake squamatesnot the postorbital The possible presence of a jugal in Dinilysiais relevant to recent arguments on the presence or absence ofa jugal in Pachyrhachis (Caldwell and Lee 1997 Zaher andRieppel 1999) and other legged snakes such as Haasiophis(Tchernov et al 2000) Zaher and Rieppel (1999) have claimedthat their studies have lsquolsquorevealedrsquorsquo the jugal of Pachyrhachis tobe a broken ectopterygoid Tchernov et al (2001) using CTScan reconstructions argue that the bone is an enlarged post-orbital that is symmetrically fractured on both sides of the ho-lotype skull While it seems unlikely that the element in ques-tion is an ectopterygoid (see the ectopterygoid of the paratypeskull illustrated by Lee and Caldwell 1998) it is intriguing thatTchernov et al (2001) would identify an element with devel-oped jugal-like topological connectivities as a postorbital sim-ilar to the condition we note in Dinilysia

Stapes and Crista Circumfenestralis As was noted by Es-tes et al (1970) Dinilysia does not possess an identifiable cristacircumfenestralis contra Rage (1984) Rieppel (1988) andTchernov et al (2000) nor is the crista prootica projecting lat-erally around the fenestra ovalis (contra Rieppel and Zaher2001) The enormous size of the stapedial footplate (Fig 3DE) may account for the absence of all cristae ie the cristacircumfenestralis if ad hoc morphogenetic explanations are in-voked (as has been argued for Acrochordus by Rieppel andZaher 2001) It may also be that Dinilysia is plesiomorphicallylacking the crista circumfenestralis as was argued by Estes etal (1970) and McDowell (1974) However at this point it canonly be stated prior to the test of congruence that the test ofsimilarity indicates that the crista circumfenestralis is absent inDinilysia

Unlike the condition in snakes (see Wever 1978) we alsofollow Estes et al (1970) by identifying the opening ventral tothe fenestra ovalis as the fenestra rotundum (Fig 2) The fe-nestra rotundum is present in most lizards but is consideredabsent in snakes and other squamates and turtles that possess areentrant fluid circuit the circuit cannot function if the roundwindow membrane is present (Wever 1978) A foramen orniche (see Wever 1978) is still present as in almost all squa-mates but in snakes it forms the channel of fluid movementfrom the recessus scala tympani to the pericapsular space Inthe otic region of Dinilysia Estes et al (1970) identified thefenestra ovalis the fenestra rotundum and the foramen for the10th cranial nerve (Fig 2D E) The difficulty with identificationof Estes et alrsquos fenestra rotundum as the lateral aperture of therecessus scala tympani is that it is not within the crista circum-fenestralis (the usual condition for extant snakes) In some basalsnakes where the crista circumfenestralis is incomplete poster-oventrally the aperture is referred to as the fenestra pseudoro-tundum

The osteology of the middle ear is also very intriguing Com-parison of the new specimens to the modern macrostomatanXenopeltis (see Frazetta 1999) indicates a number of sharedcharacters Both Dinilysia and Xenopeltis possess a broad sta-pedial footplate with a posterodorsally angled stapedial shaftThat shaft contacts a small extracolumellaintercalarystylohyalelement that articulates with the suprastapedial process of thequadrate The head of the quadrate in both taxa is very broadand laterally flattened along the contact with the supratemporalThe topographic connectivities of the supratemporal are com-

865CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 3 Postcranial skeletal elements of Dinilysia patagonica A dorsal view of articulated specimen of Dinilysia patagonica (MACN-RN-976) scale equals 10 cm B ventral view of the lsquolsquocervicalrsquorsquo series (anteriormost precloacals) of Dinilysia patagonica (MPCA-PV 527) scaleequals 1 cm CndashE anterior dorsal vertebra from holotype in lateral anterior and posterior view (MLP 26ndash410) scale equals 1 cm Abbreviationscen centrum con condyle cot cotyle d-ap diapophyses hk haemal keel na neural arch ns neural spine pactf paracotylar foramen p-apparapophyses poz postzygaophyses prz prezygapophyes przap prezygapophyseal accessory process ratna right atlas neural arch sfsubcentral foramina sg subcentral groove zgf zygantral facets zpf zygosphenal facets

mon in both snakes (ie articulating with the parietal prooticand exoccipital in a broad lsquolsquoLrsquorsquo-shaped groove) Modern aniliidsalso possess a similar osteology though the ossified intercalaryextracolumellar element is absent in Anilius (Rieppel 1980b)

Reference to the phylogenies present in the literature wouldsuggest that the absence of a crista circumfenestralis in Dini-lysia is plesiomorphic However recent descriptions of the oticregion of Achrochordus (Rieppel and Zaher 2001) would sug-gest that absence of the feature in some snake taxa may beexplained by paedomorphosis

The many new features of Dinilysia such as the presence ofa stylohyal element the absence of a crista circumfenestralisor the presence of unfused cervical intercentra raise new ques-tions about the phylogenetic relationships of snakes Questionson primitive versus derived characters for snakes center on thephylogenetic position of scolecophidians the Cretaceous ma-rine snake taxa such as Pachyrhachis and Haasiophis and earlyterrestrial Gondwanan snakes such as Dinilysia and the Madt-soiidae (Caldwell and Lee 1997 Rage and Escuille 2000Scanlon and Lee 2000 Tchernov et al 2000)

866 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ACKNOWLEDGMENTS

We thank P Posse L Salgado J Calvo and S Heredia forassistance while working and living in Neuquen We are in-debted to J Bonaparte for access to collections at the MACNand to R Pascual for permission to study Dinilysia specimensin the MLP We thank J Rose and L A Lindoe for skilledpreparation of the specimens We thank G Hanke R HolmesH Larsson A Nicholson R Nydam and S Pierce We ac-knowledge financial support from a CMN RAC Grant to MCNSERC Grant 238458-01 to MC CONICET for support of AAand a National Geographic CRE Grant for fieldwork support

LITERATURE CITED

Albino A 1996 The South American fossil Squamata (Reptilia Lep-idosauria) pp 9ndash 72 in G Arratia (ed) Contributions of SouthernSouth America to Vertebrate Paleontology Munchner Geowissenschaften Abhandlungen (A) 30

Caldwell M W 1999 Squamate phylogeny and the relationships ofsnakes and mosasauroids Zoological Journal of the Linnean So-ciety 125115ndash147

mdashmdashmdash 2000 An aquatic squamate reptile from the English ChalkDolichosaurus longicollis Owen 1850 Journal of Vertebrate Pa-leontology 20720ndash735

mdashmdashmdash and A Albino 2001 Palaeoenvironment and palaeocology ofthree Cretaceous snakes Pachyophis Pachyrhachis and DinilysiaActa Palaeontologica Polonica 46203ndash218

mdashmdashmdash and M S Y Lee 1997 A snake with legs from the marineCretaceous of the Middle East Nature 386705ndash709

Estes R T H Frazzetta and E E Williams 1970 Studies on thefossil snake Dinilysia patagonica Woodward Part 1 Cranial mor-phology Bulletin of the Museum of Comparative Zoology 14025ndash74

Frazzetta T H 1970 Studies on the fossil snake Dinilysia patagonicaWoodward Part II Jaw machinery in the earliest snakes Forma etFunctio 3205ndash221

mdashmdashmdash 1999 Adaptations and significance of the cranial feeding ap-paratus of the Sunbeam snake (Xenopeltis unicolor) Part 1 Anat-omy of the skull Journal of Morphology 23927ndash43

Hecht M 1982 The vertebral morphology of the Cretaceous snakeDinilysia patagonica Woodward Neues Jarhbuch fur Geologie undPalaontologie Monatschefte 1982523ndash532

Lee M S Y and M W Caldwell 1998 Anatomy and relationshipsof Pachyrhachis a primitive snake with hindlimbs PhilosophicalTransactions of the Royal Society Series B 3531521ndash1552

McDowell S B 1974 A catalogue of the snakes of New Guinea andthe Solomons with special reference to those in the Bernice P

Bishop museum Part 1 Scolecophidia Journal of Herpetology 81ndash57

mdashmdashmdash 1987 Systematics pp 3ndash50 in R A Siegel J T Collins andS S Novak (eds) Snakes Ecology and Evolutionary BiologyMacmillan New York

Rage J-C 1984 Serpentes Handbuch der Palaoherpetologie Teil 11Stuttgart Gustav Fischer Verlag 80 pp

mdashmdashmdash and A Albino 1989 Dinilysia patagonica (Reptilia Serpen-tes) materiel vertebral additionnel du Cretace superieurdrsquoArgentine Etude complementaire des vertebres variations in-traspecifiques et intracolumnaires Neues Jarhbuch fur Geologieund Palaontologie Monatschefte 1989433ndash447

mdashmdashmdash and F Escuillie 2000 Un nouveau serpent bipede du Ceno-manien (Cretace) Implications phyletiques Comptes Rendus delrsquoAcademie des Sciences Paris Serie Ia 3301ndash8

Rieppel O 1980a The Phylogeny of Anguinomorph Lizards BaselBirkhauser Verlag 86 pp

mdashmdashmdash 1980b The sound-transmitting apparatus in primitive snakesand its phylogenetic significance Zoomorphology 9645ndash62

mdashmdashmdash 1988 A review of the origin of snakes Evolutionary Biology2237ndash130

mdashmdashmdash and H Zaher 2001 The development of the skull in Acro-chordus granulatus (Schneider) (Reptilia Serpentes) with specialconsideration of the oticondashoccipital complex Journal of Morphol-ogy 249252ndash266

Russell D 1967 Systematics and morphology of American mosasaursPeabody Museum of Natural History Yale University Bulletin 231ndash241

Scanlon J D and M S Y Lee 2000 The Pleistocene serpent Won-ambi and the early evolution of snakes Nature 403416ndash420

Smith-Woodward A 1901 On some extinct reptiles from Patagonia ofthe genera Miolania Dinilysia and Genyodectes Proceedings ofthe Zoological Society London 1901169ndash184

Tchernov E M J Polcyn and L L Jacobs 2001 Application ofcomputed tomography to Pachyrhachis problematicus a snakewith legs from the Cretaceous of Israel Journal of Vertebrate Pa-leontology 21(3 suppl)107A

mdashmdashmdash O Rieppel H Zaher M J Polcyn and L L Jacobs 2000 Afossil snake with limbs Science 2872010ndash2012

Wever E G 1978 The Reptile Ear Its Structure and Function Prince-ton University Press Princeton 1024 pp

Zaher H and O Rieppel 1999 The phylogenetic relationships of Pa-chyrhachis problematicus and the evolution of limblessness insnakes (Lepidosauria Squamata) Comptes Rendus de lrsquoAcademiedes Sciences Paris 329831ndash837

Received 21 March 2002 accepted 17 June 2002

863CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 2 Details of skull features of Dinilysia patagonica A right lateral view of skull of MACN-RN 1013 note loose contact of maxillawith ectopterygoid postorbital and palatineprefrontal B line drawing in dorsal view of holotype (MLP 26ndash410) left maxilla postorbital andpalatineprefrontal C line drawing in ventral view of holotype (MLP 26ndash410) left maxilla and palatineprefrontal D left posteroventral viewof skull (MACN-RN 1014) emphasizing size of stapedial footplate and absence of a crista circumfenestralis E shaded line drawing of braincasein left posteroventral view Scale equals 1 cm FndashG Right ventrolateral view and line drawing of skull of MACN-RN 1014 Note preservationand articulations of stapedial shaftprocess (with suprastapedial process of enlarged quadrate) Abbreviations see Figure 1

flattens where it underlies the short robust ectopterygoid in abroad non-sutural articulation the ectopterygoid is broad andsmooth at its contact with the maxilla

The prefrontal is a small pyramid-shaped element that isbroadest along its contact with the maxilla and narrowest at itsarticulation with the frontal where the apex inserts into a shal-low v-shaped facet There is no dorsal and anterior process ofthe prefrontal (Fig 1B) Uniquely the prefrontal contributes tothe palate Along its ventral surface just anterior to the contactwith the palatine the prefrontal forms a channel that is contin-

uous with the choanal channel on the palatine (Fig 2C) andappears to be unique within squamates

The postfrontal is a small triangular bone that articulates withthe frontal and parietal and clasps the tightly sutured fronto-parietal articulation (Fig 1A B D G) the postfrontal entersthe superior margin of the orbit but does not contact the pre-frontal The postorbital is dorsoventrally elongate and shapedlike an inverted lsquolsquoTrsquorsquo The dorsal tip loosely contacts the pos-terior margin of the postfrontal and sidewall of the parietal (Fig2A) while the ventral tip broadened into a footplate-like struc-

864 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ture sits in the dorsal maxillary groove (Figs 1B 2A) Thelsquolsquojugalrsquorsquo that Estes et al (1970) identified in Dinilysia is theventral tip of the postorbital Neither a jugal nor a lachrymalcan be identified in any of the known specimens of Dinilysia

The stapedial footplate is extremely large completely ob-scures the fenestra ovalis (fenestra vestibularis) and is separat-ed from the fenestra rotundum by a low ridge of bone derivedfrom the exoccipital the crista interfenestralis (Fig 2D E) Thecolumellar shaft is at an acute angle relative to the footplateand is directed posteriorly towards the suprastapedial processof the quadrate It does not articulate directly with the supras-tapedial process but instead with a round ossified element weidentify as the extracolumella (Fig 2F G) this element has alsobeen termed the intercalary or stylohyal (for recent use of theseterms see Frazetta 1999 or Rieppel 1980a) The quadrate is alarge lsquoquestion markrsquo-shaped element with a broad cephaliccondyle robust shaft and broad mandibular condyle The su-prastapedial process is very large and expanded mediolaterallyas well as ventrally Articulation of the extracolumella-colu-mella complex is with the extreme ventromedial tip of the su-prastapedial process

Postcranial Two of the new postcranial skeletons are par-ticularly informative (Fig 3AndashE) The more complete and ar-ticulated specimen MACN RN 976 preserves 123 vertebrae ina series of fractured but associated blocks (Fig 3A) The lon-gest complete string of 40 vertebrae consists of anteriormostprecloacals that are still articulated with the skull (see Rage andAlbino 1989 for a description of the mid-precloacal vertebrae)All the vertebrae bear paracotylar foramina Unsuspected ver-tebral features are found on the newest Dinilysia specimen(MPCAndashPV 527) The third and fourth most anterior pre-cloa-cals bear large posteriorly positioned hypapophyses with largeunfused intercentra (Fig 3B) The intercentrum of the fourthprecloacal has been lost thus revealing the deeply concave sur-face of the hypapophysis (Fig 3B) The anterior hypapophysesof Dinilysia are remarkable as no known fossil or extant snakepossesses hypapophyses with unfused intercentra The posteriorposition of the hypapophyses and presence of unfused intercen-tra are also observed in mosasaurs (Russell 1967) dolichosaurs(Caldwell 2000) and varanoid lizards (Rieppel 1980b) Incontrast the fifth most-anterior precloacal bears a large hypa-pophysis likely with a fused intercentrum that is typical ofsnakes (Fig 3B)

DISCUSSION

Postfrontal and Postorbital We identify a postfrontal andpostorbital in Dinilysia While a postorbital is common to mostmodern snakes a separate postfrontal is not Estes et al (1970)identified the latter bone in Dinilysia as homologous to the post-frontal in lizards We concur with this identification as the post-frontal of Dinilysia passes the positional and relational criteriaof the test of similarity (ie element clasping the fronto-parietalsuture) and the test of composition (ie dermatocranial bone)Our identification of a postorbital in Dinilysia follows the con-vention for snakes of identifying the element that descends ven-trally from the area of the frontal-parietal suture and framingthe posterior margin of the orbit as the postorbital Though wefollow this convention for Dinilysia we do find this identifi-cation problematic due to the presence of the postfrontal andthe jugal-like articulations of the postorbital of Dinilysia

In non-snake squamates the postorbital contacts the postfron-tal sometimes fuses with it extends a posterior ramus to con-tact the squamosal and contacts the jugal where that elementrises to meet the posterior margin of the postfrontal In contrastthe jugal bone variably articulates with the postfrontal postor-bital or postorbital frontal articulates with the maxilla behindand beneath the orbit and articulates with the lachrymal if pre-

sent The jugal bone of the varanoid lizard Lanthanotus artic-ulates with the postfrontal maxilla-ectopterygoid and lachry-mal (the postorbital is absent in Lanthanotus) The remainingarticulation ie with the ectopterygoid is variable among liz-ards and snakes (eg in Python the lsquolsquopostorbitalrsquorsquo contacts theectopterygoid) Comparisons to snakes such as scolecophidiansare not possible as all superior and posterior orbital and tem-poral-arch bones are absent The test of topological relationsuggests a closer lsquofitrsquo with the jugal of non-snake squamatesnot the postorbital The possible presence of a jugal in Dinilysiais relevant to recent arguments on the presence or absence ofa jugal in Pachyrhachis (Caldwell and Lee 1997 Zaher andRieppel 1999) and other legged snakes such as Haasiophis(Tchernov et al 2000) Zaher and Rieppel (1999) have claimedthat their studies have lsquolsquorevealedrsquorsquo the jugal of Pachyrhachis tobe a broken ectopterygoid Tchernov et al (2001) using CTScan reconstructions argue that the bone is an enlarged post-orbital that is symmetrically fractured on both sides of the ho-lotype skull While it seems unlikely that the element in ques-tion is an ectopterygoid (see the ectopterygoid of the paratypeskull illustrated by Lee and Caldwell 1998) it is intriguing thatTchernov et al (2001) would identify an element with devel-oped jugal-like topological connectivities as a postorbital sim-ilar to the condition we note in Dinilysia

Stapes and Crista Circumfenestralis As was noted by Es-tes et al (1970) Dinilysia does not possess an identifiable cristacircumfenestralis contra Rage (1984) Rieppel (1988) andTchernov et al (2000) nor is the crista prootica projecting lat-erally around the fenestra ovalis (contra Rieppel and Zaher2001) The enormous size of the stapedial footplate (Fig 3DE) may account for the absence of all cristae ie the cristacircumfenestralis if ad hoc morphogenetic explanations are in-voked (as has been argued for Acrochordus by Rieppel andZaher 2001) It may also be that Dinilysia is plesiomorphicallylacking the crista circumfenestralis as was argued by Estes etal (1970) and McDowell (1974) However at this point it canonly be stated prior to the test of congruence that the test ofsimilarity indicates that the crista circumfenestralis is absent inDinilysia

Unlike the condition in snakes (see Wever 1978) we alsofollow Estes et al (1970) by identifying the opening ventral tothe fenestra ovalis as the fenestra rotundum (Fig 2) The fe-nestra rotundum is present in most lizards but is consideredabsent in snakes and other squamates and turtles that possess areentrant fluid circuit the circuit cannot function if the roundwindow membrane is present (Wever 1978) A foramen orniche (see Wever 1978) is still present as in almost all squa-mates but in snakes it forms the channel of fluid movementfrom the recessus scala tympani to the pericapsular space Inthe otic region of Dinilysia Estes et al (1970) identified thefenestra ovalis the fenestra rotundum and the foramen for the10th cranial nerve (Fig 2D E) The difficulty with identificationof Estes et alrsquos fenestra rotundum as the lateral aperture of therecessus scala tympani is that it is not within the crista circum-fenestralis (the usual condition for extant snakes) In some basalsnakes where the crista circumfenestralis is incomplete poster-oventrally the aperture is referred to as the fenestra pseudoro-tundum

The osteology of the middle ear is also very intriguing Com-parison of the new specimens to the modern macrostomatanXenopeltis (see Frazetta 1999) indicates a number of sharedcharacters Both Dinilysia and Xenopeltis possess a broad sta-pedial footplate with a posterodorsally angled stapedial shaftThat shaft contacts a small extracolumellaintercalarystylohyalelement that articulates with the suprastapedial process of thequadrate The head of the quadrate in both taxa is very broadand laterally flattened along the contact with the supratemporalThe topographic connectivities of the supratemporal are com-

865CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 3 Postcranial skeletal elements of Dinilysia patagonica A dorsal view of articulated specimen of Dinilysia patagonica (MACN-RN-976) scale equals 10 cm B ventral view of the lsquolsquocervicalrsquorsquo series (anteriormost precloacals) of Dinilysia patagonica (MPCA-PV 527) scaleequals 1 cm CndashE anterior dorsal vertebra from holotype in lateral anterior and posterior view (MLP 26ndash410) scale equals 1 cm Abbreviationscen centrum con condyle cot cotyle d-ap diapophyses hk haemal keel na neural arch ns neural spine pactf paracotylar foramen p-apparapophyses poz postzygaophyses prz prezygapophyes przap prezygapophyseal accessory process ratna right atlas neural arch sfsubcentral foramina sg subcentral groove zgf zygantral facets zpf zygosphenal facets

mon in both snakes (ie articulating with the parietal prooticand exoccipital in a broad lsquolsquoLrsquorsquo-shaped groove) Modern aniliidsalso possess a similar osteology though the ossified intercalaryextracolumellar element is absent in Anilius (Rieppel 1980b)

Reference to the phylogenies present in the literature wouldsuggest that the absence of a crista circumfenestralis in Dini-lysia is plesiomorphic However recent descriptions of the oticregion of Achrochordus (Rieppel and Zaher 2001) would sug-gest that absence of the feature in some snake taxa may beexplained by paedomorphosis

The many new features of Dinilysia such as the presence ofa stylohyal element the absence of a crista circumfenestralisor the presence of unfused cervical intercentra raise new ques-tions about the phylogenetic relationships of snakes Questionson primitive versus derived characters for snakes center on thephylogenetic position of scolecophidians the Cretaceous ma-rine snake taxa such as Pachyrhachis and Haasiophis and earlyterrestrial Gondwanan snakes such as Dinilysia and the Madt-soiidae (Caldwell and Lee 1997 Rage and Escuille 2000Scanlon and Lee 2000 Tchernov et al 2000)

866 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ACKNOWLEDGMENTS

We thank P Posse L Salgado J Calvo and S Heredia forassistance while working and living in Neuquen We are in-debted to J Bonaparte for access to collections at the MACNand to R Pascual for permission to study Dinilysia specimensin the MLP We thank J Rose and L A Lindoe for skilledpreparation of the specimens We thank G Hanke R HolmesH Larsson A Nicholson R Nydam and S Pierce We ac-knowledge financial support from a CMN RAC Grant to MCNSERC Grant 238458-01 to MC CONICET for support of AAand a National Geographic CRE Grant for fieldwork support

LITERATURE CITED

Albino A 1996 The South American fossil Squamata (Reptilia Lep-idosauria) pp 9ndash 72 in G Arratia (ed) Contributions of SouthernSouth America to Vertebrate Paleontology Munchner Geowissenschaften Abhandlungen (A) 30

Caldwell M W 1999 Squamate phylogeny and the relationships ofsnakes and mosasauroids Zoological Journal of the Linnean So-ciety 125115ndash147

mdashmdashmdash 2000 An aquatic squamate reptile from the English ChalkDolichosaurus longicollis Owen 1850 Journal of Vertebrate Pa-leontology 20720ndash735

mdashmdashmdash and A Albino 2001 Palaeoenvironment and palaeocology ofthree Cretaceous snakes Pachyophis Pachyrhachis and DinilysiaActa Palaeontologica Polonica 46203ndash218

mdashmdashmdash and M S Y Lee 1997 A snake with legs from the marineCretaceous of the Middle East Nature 386705ndash709

Estes R T H Frazzetta and E E Williams 1970 Studies on thefossil snake Dinilysia patagonica Woodward Part 1 Cranial mor-phology Bulletin of the Museum of Comparative Zoology 14025ndash74

Frazzetta T H 1970 Studies on the fossil snake Dinilysia patagonicaWoodward Part II Jaw machinery in the earliest snakes Forma etFunctio 3205ndash221

mdashmdashmdash 1999 Adaptations and significance of the cranial feeding ap-paratus of the Sunbeam snake (Xenopeltis unicolor) Part 1 Anat-omy of the skull Journal of Morphology 23927ndash43

Hecht M 1982 The vertebral morphology of the Cretaceous snakeDinilysia patagonica Woodward Neues Jarhbuch fur Geologie undPalaontologie Monatschefte 1982523ndash532

Lee M S Y and M W Caldwell 1998 Anatomy and relationshipsof Pachyrhachis a primitive snake with hindlimbs PhilosophicalTransactions of the Royal Society Series B 3531521ndash1552

McDowell S B 1974 A catalogue of the snakes of New Guinea andthe Solomons with special reference to those in the Bernice P

Bishop museum Part 1 Scolecophidia Journal of Herpetology 81ndash57

mdashmdashmdash 1987 Systematics pp 3ndash50 in R A Siegel J T Collins andS S Novak (eds) Snakes Ecology and Evolutionary BiologyMacmillan New York

Rage J-C 1984 Serpentes Handbuch der Palaoherpetologie Teil 11Stuttgart Gustav Fischer Verlag 80 pp

mdashmdashmdash and A Albino 1989 Dinilysia patagonica (Reptilia Serpen-tes) materiel vertebral additionnel du Cretace superieurdrsquoArgentine Etude complementaire des vertebres variations in-traspecifiques et intracolumnaires Neues Jarhbuch fur Geologieund Palaontologie Monatschefte 1989433ndash447

mdashmdashmdash and F Escuillie 2000 Un nouveau serpent bipede du Ceno-manien (Cretace) Implications phyletiques Comptes Rendus delrsquoAcademie des Sciences Paris Serie Ia 3301ndash8

Rieppel O 1980a The Phylogeny of Anguinomorph Lizards BaselBirkhauser Verlag 86 pp

mdashmdashmdash 1980b The sound-transmitting apparatus in primitive snakesand its phylogenetic significance Zoomorphology 9645ndash62

mdashmdashmdash 1988 A review of the origin of snakes Evolutionary Biology2237ndash130

mdashmdashmdash and H Zaher 2001 The development of the skull in Acro-chordus granulatus (Schneider) (Reptilia Serpentes) with specialconsideration of the oticondashoccipital complex Journal of Morphol-ogy 249252ndash266

Russell D 1967 Systematics and morphology of American mosasaursPeabody Museum of Natural History Yale University Bulletin 231ndash241

Scanlon J D and M S Y Lee 2000 The Pleistocene serpent Won-ambi and the early evolution of snakes Nature 403416ndash420

Smith-Woodward A 1901 On some extinct reptiles from Patagonia ofthe genera Miolania Dinilysia and Genyodectes Proceedings ofthe Zoological Society London 1901169ndash184

Tchernov E M J Polcyn and L L Jacobs 2001 Application ofcomputed tomography to Pachyrhachis problematicus a snakewith legs from the Cretaceous of Israel Journal of Vertebrate Pa-leontology 21(3 suppl)107A

mdashmdashmdash O Rieppel H Zaher M J Polcyn and L L Jacobs 2000 Afossil snake with limbs Science 2872010ndash2012

Wever E G 1978 The Reptile Ear Its Structure and Function Prince-ton University Press Princeton 1024 pp

Zaher H and O Rieppel 1999 The phylogenetic relationships of Pa-chyrhachis problematicus and the evolution of limblessness insnakes (Lepidosauria Squamata) Comptes Rendus de lrsquoAcademiedes Sciences Paris 329831ndash837

Received 21 March 2002 accepted 17 June 2002

864 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ture sits in the dorsal maxillary groove (Figs 1B 2A) Thelsquolsquojugalrsquorsquo that Estes et al (1970) identified in Dinilysia is theventral tip of the postorbital Neither a jugal nor a lachrymalcan be identified in any of the known specimens of Dinilysia

The stapedial footplate is extremely large completely ob-scures the fenestra ovalis (fenestra vestibularis) and is separat-ed from the fenestra rotundum by a low ridge of bone derivedfrom the exoccipital the crista interfenestralis (Fig 2D E) Thecolumellar shaft is at an acute angle relative to the footplateand is directed posteriorly towards the suprastapedial processof the quadrate It does not articulate directly with the supras-tapedial process but instead with a round ossified element weidentify as the extracolumella (Fig 2F G) this element has alsobeen termed the intercalary or stylohyal (for recent use of theseterms see Frazetta 1999 or Rieppel 1980a) The quadrate is alarge lsquoquestion markrsquo-shaped element with a broad cephaliccondyle robust shaft and broad mandibular condyle The su-prastapedial process is very large and expanded mediolaterallyas well as ventrally Articulation of the extracolumella-colu-mella complex is with the extreme ventromedial tip of the su-prastapedial process

Postcranial Two of the new postcranial skeletons are par-ticularly informative (Fig 3AndashE) The more complete and ar-ticulated specimen MACN RN 976 preserves 123 vertebrae ina series of fractured but associated blocks (Fig 3A) The lon-gest complete string of 40 vertebrae consists of anteriormostprecloacals that are still articulated with the skull (see Rage andAlbino 1989 for a description of the mid-precloacal vertebrae)All the vertebrae bear paracotylar foramina Unsuspected ver-tebral features are found on the newest Dinilysia specimen(MPCAndashPV 527) The third and fourth most anterior pre-cloa-cals bear large posteriorly positioned hypapophyses with largeunfused intercentra (Fig 3B) The intercentrum of the fourthprecloacal has been lost thus revealing the deeply concave sur-face of the hypapophysis (Fig 3B) The anterior hypapophysesof Dinilysia are remarkable as no known fossil or extant snakepossesses hypapophyses with unfused intercentra The posteriorposition of the hypapophyses and presence of unfused intercen-tra are also observed in mosasaurs (Russell 1967) dolichosaurs(Caldwell 2000) and varanoid lizards (Rieppel 1980b) Incontrast the fifth most-anterior precloacal bears a large hypa-pophysis likely with a fused intercentrum that is typical ofsnakes (Fig 3B)

DISCUSSION

Postfrontal and Postorbital We identify a postfrontal andpostorbital in Dinilysia While a postorbital is common to mostmodern snakes a separate postfrontal is not Estes et al (1970)identified the latter bone in Dinilysia as homologous to the post-frontal in lizards We concur with this identification as the post-frontal of Dinilysia passes the positional and relational criteriaof the test of similarity (ie element clasping the fronto-parietalsuture) and the test of composition (ie dermatocranial bone)Our identification of a postorbital in Dinilysia follows the con-vention for snakes of identifying the element that descends ven-trally from the area of the frontal-parietal suture and framingthe posterior margin of the orbit as the postorbital Though wefollow this convention for Dinilysia we do find this identifi-cation problematic due to the presence of the postfrontal andthe jugal-like articulations of the postorbital of Dinilysia

In non-snake squamates the postorbital contacts the postfron-tal sometimes fuses with it extends a posterior ramus to con-tact the squamosal and contacts the jugal where that elementrises to meet the posterior margin of the postfrontal In contrastthe jugal bone variably articulates with the postfrontal postor-bital or postorbital frontal articulates with the maxilla behindand beneath the orbit and articulates with the lachrymal if pre-

sent The jugal bone of the varanoid lizard Lanthanotus artic-ulates with the postfrontal maxilla-ectopterygoid and lachry-mal (the postorbital is absent in Lanthanotus) The remainingarticulation ie with the ectopterygoid is variable among liz-ards and snakes (eg in Python the lsquolsquopostorbitalrsquorsquo contacts theectopterygoid) Comparisons to snakes such as scolecophidiansare not possible as all superior and posterior orbital and tem-poral-arch bones are absent The test of topological relationsuggests a closer lsquofitrsquo with the jugal of non-snake squamatesnot the postorbital The possible presence of a jugal in Dinilysiais relevant to recent arguments on the presence or absence ofa jugal in Pachyrhachis (Caldwell and Lee 1997 Zaher andRieppel 1999) and other legged snakes such as Haasiophis(Tchernov et al 2000) Zaher and Rieppel (1999) have claimedthat their studies have lsquolsquorevealedrsquorsquo the jugal of Pachyrhachis tobe a broken ectopterygoid Tchernov et al (2001) using CTScan reconstructions argue that the bone is an enlarged post-orbital that is symmetrically fractured on both sides of the ho-lotype skull While it seems unlikely that the element in ques-tion is an ectopterygoid (see the ectopterygoid of the paratypeskull illustrated by Lee and Caldwell 1998) it is intriguing thatTchernov et al (2001) would identify an element with devel-oped jugal-like topological connectivities as a postorbital sim-ilar to the condition we note in Dinilysia

Stapes and Crista Circumfenestralis As was noted by Es-tes et al (1970) Dinilysia does not possess an identifiable cristacircumfenestralis contra Rage (1984) Rieppel (1988) andTchernov et al (2000) nor is the crista prootica projecting lat-erally around the fenestra ovalis (contra Rieppel and Zaher2001) The enormous size of the stapedial footplate (Fig 3DE) may account for the absence of all cristae ie the cristacircumfenestralis if ad hoc morphogenetic explanations are in-voked (as has been argued for Acrochordus by Rieppel andZaher 2001) It may also be that Dinilysia is plesiomorphicallylacking the crista circumfenestralis as was argued by Estes etal (1970) and McDowell (1974) However at this point it canonly be stated prior to the test of congruence that the test ofsimilarity indicates that the crista circumfenestralis is absent inDinilysia

Unlike the condition in snakes (see Wever 1978) we alsofollow Estes et al (1970) by identifying the opening ventral tothe fenestra ovalis as the fenestra rotundum (Fig 2) The fe-nestra rotundum is present in most lizards but is consideredabsent in snakes and other squamates and turtles that possess areentrant fluid circuit the circuit cannot function if the roundwindow membrane is present (Wever 1978) A foramen orniche (see Wever 1978) is still present as in almost all squa-mates but in snakes it forms the channel of fluid movementfrom the recessus scala tympani to the pericapsular space Inthe otic region of Dinilysia Estes et al (1970) identified thefenestra ovalis the fenestra rotundum and the foramen for the10th cranial nerve (Fig 2D E) The difficulty with identificationof Estes et alrsquos fenestra rotundum as the lateral aperture of therecessus scala tympani is that it is not within the crista circum-fenestralis (the usual condition for extant snakes) In some basalsnakes where the crista circumfenestralis is incomplete poster-oventrally the aperture is referred to as the fenestra pseudoro-tundum

The osteology of the middle ear is also very intriguing Com-parison of the new specimens to the modern macrostomatanXenopeltis (see Frazetta 1999) indicates a number of sharedcharacters Both Dinilysia and Xenopeltis possess a broad sta-pedial footplate with a posterodorsally angled stapedial shaftThat shaft contacts a small extracolumellaintercalarystylohyalelement that articulates with the suprastapedial process of thequadrate The head of the quadrate in both taxa is very broadand laterally flattened along the contact with the supratemporalThe topographic connectivities of the supratemporal are com-

865CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 3 Postcranial skeletal elements of Dinilysia patagonica A dorsal view of articulated specimen of Dinilysia patagonica (MACN-RN-976) scale equals 10 cm B ventral view of the lsquolsquocervicalrsquorsquo series (anteriormost precloacals) of Dinilysia patagonica (MPCA-PV 527) scaleequals 1 cm CndashE anterior dorsal vertebra from holotype in lateral anterior and posterior view (MLP 26ndash410) scale equals 1 cm Abbreviationscen centrum con condyle cot cotyle d-ap diapophyses hk haemal keel na neural arch ns neural spine pactf paracotylar foramen p-apparapophyses poz postzygaophyses prz prezygapophyes przap prezygapophyseal accessory process ratna right atlas neural arch sfsubcentral foramina sg subcentral groove zgf zygantral facets zpf zygosphenal facets

mon in both snakes (ie articulating with the parietal prooticand exoccipital in a broad lsquolsquoLrsquorsquo-shaped groove) Modern aniliidsalso possess a similar osteology though the ossified intercalaryextracolumellar element is absent in Anilius (Rieppel 1980b)

Reference to the phylogenies present in the literature wouldsuggest that the absence of a crista circumfenestralis in Dini-lysia is plesiomorphic However recent descriptions of the oticregion of Achrochordus (Rieppel and Zaher 2001) would sug-gest that absence of the feature in some snake taxa may beexplained by paedomorphosis

The many new features of Dinilysia such as the presence ofa stylohyal element the absence of a crista circumfenestralisor the presence of unfused cervical intercentra raise new ques-tions about the phylogenetic relationships of snakes Questionson primitive versus derived characters for snakes center on thephylogenetic position of scolecophidians the Cretaceous ma-rine snake taxa such as Pachyrhachis and Haasiophis and earlyterrestrial Gondwanan snakes such as Dinilysia and the Madt-soiidae (Caldwell and Lee 1997 Rage and Escuille 2000Scanlon and Lee 2000 Tchernov et al 2000)

866 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ACKNOWLEDGMENTS

We thank P Posse L Salgado J Calvo and S Heredia forassistance while working and living in Neuquen We are in-debted to J Bonaparte for access to collections at the MACNand to R Pascual for permission to study Dinilysia specimensin the MLP We thank J Rose and L A Lindoe for skilledpreparation of the specimens We thank G Hanke R HolmesH Larsson A Nicholson R Nydam and S Pierce We ac-knowledge financial support from a CMN RAC Grant to MCNSERC Grant 238458-01 to MC CONICET for support of AAand a National Geographic CRE Grant for fieldwork support

LITERATURE CITED

Albino A 1996 The South American fossil Squamata (Reptilia Lep-idosauria) pp 9ndash 72 in G Arratia (ed) Contributions of SouthernSouth America to Vertebrate Paleontology Munchner Geowissenschaften Abhandlungen (A) 30

Caldwell M W 1999 Squamate phylogeny and the relationships ofsnakes and mosasauroids Zoological Journal of the Linnean So-ciety 125115ndash147

mdashmdashmdash 2000 An aquatic squamate reptile from the English ChalkDolichosaurus longicollis Owen 1850 Journal of Vertebrate Pa-leontology 20720ndash735

mdashmdashmdash and A Albino 2001 Palaeoenvironment and palaeocology ofthree Cretaceous snakes Pachyophis Pachyrhachis and DinilysiaActa Palaeontologica Polonica 46203ndash218

mdashmdashmdash and M S Y Lee 1997 A snake with legs from the marineCretaceous of the Middle East Nature 386705ndash709

Estes R T H Frazzetta and E E Williams 1970 Studies on thefossil snake Dinilysia patagonica Woodward Part 1 Cranial mor-phology Bulletin of the Museum of Comparative Zoology 14025ndash74

Frazzetta T H 1970 Studies on the fossil snake Dinilysia patagonicaWoodward Part II Jaw machinery in the earliest snakes Forma etFunctio 3205ndash221

mdashmdashmdash 1999 Adaptations and significance of the cranial feeding ap-paratus of the Sunbeam snake (Xenopeltis unicolor) Part 1 Anat-omy of the skull Journal of Morphology 23927ndash43

Hecht M 1982 The vertebral morphology of the Cretaceous snakeDinilysia patagonica Woodward Neues Jarhbuch fur Geologie undPalaontologie Monatschefte 1982523ndash532

Lee M S Y and M W Caldwell 1998 Anatomy and relationshipsof Pachyrhachis a primitive snake with hindlimbs PhilosophicalTransactions of the Royal Society Series B 3531521ndash1552

McDowell S B 1974 A catalogue of the snakes of New Guinea andthe Solomons with special reference to those in the Bernice P

Bishop museum Part 1 Scolecophidia Journal of Herpetology 81ndash57

mdashmdashmdash 1987 Systematics pp 3ndash50 in R A Siegel J T Collins andS S Novak (eds) Snakes Ecology and Evolutionary BiologyMacmillan New York

Rage J-C 1984 Serpentes Handbuch der Palaoherpetologie Teil 11Stuttgart Gustav Fischer Verlag 80 pp

mdashmdashmdash and A Albino 1989 Dinilysia patagonica (Reptilia Serpen-tes) materiel vertebral additionnel du Cretace superieurdrsquoArgentine Etude complementaire des vertebres variations in-traspecifiques et intracolumnaires Neues Jarhbuch fur Geologieund Palaontologie Monatschefte 1989433ndash447

mdashmdashmdash and F Escuillie 2000 Un nouveau serpent bipede du Ceno-manien (Cretace) Implications phyletiques Comptes Rendus delrsquoAcademie des Sciences Paris Serie Ia 3301ndash8

Rieppel O 1980a The Phylogeny of Anguinomorph Lizards BaselBirkhauser Verlag 86 pp

mdashmdashmdash 1980b The sound-transmitting apparatus in primitive snakesand its phylogenetic significance Zoomorphology 9645ndash62

mdashmdashmdash 1988 A review of the origin of snakes Evolutionary Biology2237ndash130

mdashmdashmdash and H Zaher 2001 The development of the skull in Acro-chordus granulatus (Schneider) (Reptilia Serpentes) with specialconsideration of the oticondashoccipital complex Journal of Morphol-ogy 249252ndash266

Russell D 1967 Systematics and morphology of American mosasaursPeabody Museum of Natural History Yale University Bulletin 231ndash241

Scanlon J D and M S Y Lee 2000 The Pleistocene serpent Won-ambi and the early evolution of snakes Nature 403416ndash420

Smith-Woodward A 1901 On some extinct reptiles from Patagonia ofthe genera Miolania Dinilysia and Genyodectes Proceedings ofthe Zoological Society London 1901169ndash184

Tchernov E M J Polcyn and L L Jacobs 2001 Application ofcomputed tomography to Pachyrhachis problematicus a snakewith legs from the Cretaceous of Israel Journal of Vertebrate Pa-leontology 21(3 suppl)107A

mdashmdashmdash O Rieppel H Zaher M J Polcyn and L L Jacobs 2000 Afossil snake with limbs Science 2872010ndash2012

Wever E G 1978 The Reptile Ear Its Structure and Function Prince-ton University Press Princeton 1024 pp

Zaher H and O Rieppel 1999 The phylogenetic relationships of Pa-chyrhachis problematicus and the evolution of limblessness insnakes (Lepidosauria Squamata) Comptes Rendus de lrsquoAcademiedes Sciences Paris 329831ndash837

Received 21 March 2002 accepted 17 June 2002

865CALDWELL AND ALBINOmdashCRETACEOUS SNAKE DINILYSIA

FIGURE 3 Postcranial skeletal elements of Dinilysia patagonica A dorsal view of articulated specimen of Dinilysia patagonica (MACN-RN-976) scale equals 10 cm B ventral view of the lsquolsquocervicalrsquorsquo series (anteriormost precloacals) of Dinilysia patagonica (MPCA-PV 527) scaleequals 1 cm CndashE anterior dorsal vertebra from holotype in lateral anterior and posterior view (MLP 26ndash410) scale equals 1 cm Abbreviationscen centrum con condyle cot cotyle d-ap diapophyses hk haemal keel na neural arch ns neural spine pactf paracotylar foramen p-apparapophyses poz postzygaophyses prz prezygapophyes przap prezygapophyseal accessory process ratna right atlas neural arch sfsubcentral foramina sg subcentral groove zgf zygantral facets zpf zygosphenal facets

mon in both snakes (ie articulating with the parietal prooticand exoccipital in a broad lsquolsquoLrsquorsquo-shaped groove) Modern aniliidsalso possess a similar osteology though the ossified intercalaryextracolumellar element is absent in Anilius (Rieppel 1980b)

Reference to the phylogenies present in the literature wouldsuggest that the absence of a crista circumfenestralis in Dini-lysia is plesiomorphic However recent descriptions of the oticregion of Achrochordus (Rieppel and Zaher 2001) would sug-gest that absence of the feature in some snake taxa may beexplained by paedomorphosis

The many new features of Dinilysia such as the presence ofa stylohyal element the absence of a crista circumfenestralisor the presence of unfused cervical intercentra raise new ques-tions about the phylogenetic relationships of snakes Questionson primitive versus derived characters for snakes center on thephylogenetic position of scolecophidians the Cretaceous ma-rine snake taxa such as Pachyrhachis and Haasiophis and earlyterrestrial Gondwanan snakes such as Dinilysia and the Madt-soiidae (Caldwell and Lee 1997 Rage and Escuille 2000Scanlon and Lee 2000 Tchernov et al 2000)

866 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ACKNOWLEDGMENTS

We thank P Posse L Salgado J Calvo and S Heredia forassistance while working and living in Neuquen We are in-debted to J Bonaparte for access to collections at the MACNand to R Pascual for permission to study Dinilysia specimensin the MLP We thank J Rose and L A Lindoe for skilledpreparation of the specimens We thank G Hanke R HolmesH Larsson A Nicholson R Nydam and S Pierce We ac-knowledge financial support from a CMN RAC Grant to MCNSERC Grant 238458-01 to MC CONICET for support of AAand a National Geographic CRE Grant for fieldwork support

LITERATURE CITED

Albino A 1996 The South American fossil Squamata (Reptilia Lep-idosauria) pp 9ndash 72 in G Arratia (ed) Contributions of SouthernSouth America to Vertebrate Paleontology Munchner Geowissenschaften Abhandlungen (A) 30

Caldwell M W 1999 Squamate phylogeny and the relationships ofsnakes and mosasauroids Zoological Journal of the Linnean So-ciety 125115ndash147

mdashmdashmdash 2000 An aquatic squamate reptile from the English ChalkDolichosaurus longicollis Owen 1850 Journal of Vertebrate Pa-leontology 20720ndash735

mdashmdashmdash and A Albino 2001 Palaeoenvironment and palaeocology ofthree Cretaceous snakes Pachyophis Pachyrhachis and DinilysiaActa Palaeontologica Polonica 46203ndash218

mdashmdashmdash and M S Y Lee 1997 A snake with legs from the marineCretaceous of the Middle East Nature 386705ndash709

Estes R T H Frazzetta and E E Williams 1970 Studies on thefossil snake Dinilysia patagonica Woodward Part 1 Cranial mor-phology Bulletin of the Museum of Comparative Zoology 14025ndash74

Frazzetta T H 1970 Studies on the fossil snake Dinilysia patagonicaWoodward Part II Jaw machinery in the earliest snakes Forma etFunctio 3205ndash221

mdashmdashmdash 1999 Adaptations and significance of the cranial feeding ap-paratus of the Sunbeam snake (Xenopeltis unicolor) Part 1 Anat-omy of the skull Journal of Morphology 23927ndash43

Hecht M 1982 The vertebral morphology of the Cretaceous snakeDinilysia patagonica Woodward Neues Jarhbuch fur Geologie undPalaontologie Monatschefte 1982523ndash532

Lee M S Y and M W Caldwell 1998 Anatomy and relationshipsof Pachyrhachis a primitive snake with hindlimbs PhilosophicalTransactions of the Royal Society Series B 3531521ndash1552

McDowell S B 1974 A catalogue of the snakes of New Guinea andthe Solomons with special reference to those in the Bernice P

Bishop museum Part 1 Scolecophidia Journal of Herpetology 81ndash57

mdashmdashmdash 1987 Systematics pp 3ndash50 in R A Siegel J T Collins andS S Novak (eds) Snakes Ecology and Evolutionary BiologyMacmillan New York

Rage J-C 1984 Serpentes Handbuch der Palaoherpetologie Teil 11Stuttgart Gustav Fischer Verlag 80 pp

mdashmdashmdash and A Albino 1989 Dinilysia patagonica (Reptilia Serpen-tes) materiel vertebral additionnel du Cretace superieurdrsquoArgentine Etude complementaire des vertebres variations in-traspecifiques et intracolumnaires Neues Jarhbuch fur Geologieund Palaontologie Monatschefte 1989433ndash447

mdashmdashmdash and F Escuillie 2000 Un nouveau serpent bipede du Ceno-manien (Cretace) Implications phyletiques Comptes Rendus delrsquoAcademie des Sciences Paris Serie Ia 3301ndash8

Rieppel O 1980a The Phylogeny of Anguinomorph Lizards BaselBirkhauser Verlag 86 pp

mdashmdashmdash 1980b The sound-transmitting apparatus in primitive snakesand its phylogenetic significance Zoomorphology 9645ndash62

mdashmdashmdash 1988 A review of the origin of snakes Evolutionary Biology2237ndash130

mdashmdashmdash and H Zaher 2001 The development of the skull in Acro-chordus granulatus (Schneider) (Reptilia Serpentes) with specialconsideration of the oticondashoccipital complex Journal of Morphol-ogy 249252ndash266

Russell D 1967 Systematics and morphology of American mosasaursPeabody Museum of Natural History Yale University Bulletin 231ndash241

Scanlon J D and M S Y Lee 2000 The Pleistocene serpent Won-ambi and the early evolution of snakes Nature 403416ndash420

Smith-Woodward A 1901 On some extinct reptiles from Patagonia ofthe genera Miolania Dinilysia and Genyodectes Proceedings ofthe Zoological Society London 1901169ndash184

Tchernov E M J Polcyn and L L Jacobs 2001 Application ofcomputed tomography to Pachyrhachis problematicus a snakewith legs from the Cretaceous of Israel Journal of Vertebrate Pa-leontology 21(3 suppl)107A

mdashmdashmdash O Rieppel H Zaher M J Polcyn and L L Jacobs 2000 Afossil snake with limbs Science 2872010ndash2012

Wever E G 1978 The Reptile Ear Its Structure and Function Prince-ton University Press Princeton 1024 pp

Zaher H and O Rieppel 1999 The phylogenetic relationships of Pa-chyrhachis problematicus and the evolution of limblessness insnakes (Lepidosauria Squamata) Comptes Rendus de lrsquoAcademiedes Sciences Paris 329831ndash837

Received 21 March 2002 accepted 17 June 2002

866 JOURNAL OF VERTEBRATE PALEONTOLOGY VOL 22 NO 4 2002

ACKNOWLEDGMENTS

We thank P Posse L Salgado J Calvo and S Heredia forassistance while working and living in Neuquen We are in-debted to J Bonaparte for access to collections at the MACNand to R Pascual for permission to study Dinilysia specimensin the MLP We thank J Rose and L A Lindoe for skilledpreparation of the specimens We thank G Hanke R HolmesH Larsson A Nicholson R Nydam and S Pierce We ac-knowledge financial support from a CMN RAC Grant to MCNSERC Grant 238458-01 to MC CONICET for support of AAand a National Geographic CRE Grant for fieldwork support

LITERATURE CITED

Albino A 1996 The South American fossil Squamata (Reptilia Lep-idosauria) pp 9ndash 72 in G Arratia (ed) Contributions of SouthernSouth America to Vertebrate Paleontology Munchner Geowissenschaften Abhandlungen (A) 30

Caldwell M W 1999 Squamate phylogeny and the relationships ofsnakes and mosasauroids Zoological Journal of the Linnean So-ciety 125115ndash147

mdashmdashmdash 2000 An aquatic squamate reptile from the English ChalkDolichosaurus longicollis Owen 1850 Journal of Vertebrate Pa-leontology 20720ndash735

mdashmdashmdash and A Albino 2001 Palaeoenvironment and palaeocology ofthree Cretaceous snakes Pachyophis Pachyrhachis and DinilysiaActa Palaeontologica Polonica 46203ndash218

mdashmdashmdash and M S Y Lee 1997 A snake with legs from the marineCretaceous of the Middle East Nature 386705ndash709

Estes R T H Frazzetta and E E Williams 1970 Studies on thefossil snake Dinilysia patagonica Woodward Part 1 Cranial mor-phology Bulletin of the Museum of Comparative Zoology 14025ndash74

Frazzetta T H 1970 Studies on the fossil snake Dinilysia patagonicaWoodward Part II Jaw machinery in the earliest snakes Forma etFunctio 3205ndash221

mdashmdashmdash 1999 Adaptations and significance of the cranial feeding ap-paratus of the Sunbeam snake (Xenopeltis unicolor) Part 1 Anat-omy of the skull Journal of Morphology 23927ndash43

Hecht M 1982 The vertebral morphology of the Cretaceous snakeDinilysia patagonica Woodward Neues Jarhbuch fur Geologie undPalaontologie Monatschefte 1982523ndash532

Lee M S Y and M W Caldwell 1998 Anatomy and relationshipsof Pachyrhachis a primitive snake with hindlimbs PhilosophicalTransactions of the Royal Society Series B 3531521ndash1552

McDowell S B 1974 A catalogue of the snakes of New Guinea andthe Solomons with special reference to those in the Bernice P

Bishop museum Part 1 Scolecophidia Journal of Herpetology 81ndash57

mdashmdashmdash 1987 Systematics pp 3ndash50 in R A Siegel J T Collins andS S Novak (eds) Snakes Ecology and Evolutionary BiologyMacmillan New York

Rage J-C 1984 Serpentes Handbuch der Palaoherpetologie Teil 11Stuttgart Gustav Fischer Verlag 80 pp

mdashmdashmdash and A Albino 1989 Dinilysia patagonica (Reptilia Serpen-tes) materiel vertebral additionnel du Cretace superieurdrsquoArgentine Etude complementaire des vertebres variations in-traspecifiques et intracolumnaires Neues Jarhbuch fur Geologieund Palaontologie Monatschefte 1989433ndash447

mdashmdashmdash and F Escuillie 2000 Un nouveau serpent bipede du Ceno-manien (Cretace) Implications phyletiques Comptes Rendus delrsquoAcademie des Sciences Paris Serie Ia 3301ndash8

Rieppel O 1980a The Phylogeny of Anguinomorph Lizards BaselBirkhauser Verlag 86 pp

mdashmdashmdash 1980b The sound-transmitting apparatus in primitive snakesand its phylogenetic significance Zoomorphology 9645ndash62

mdashmdashmdash 1988 A review of the origin of snakes Evolutionary Biology2237ndash130

mdashmdashmdash and H Zaher 2001 The development of the skull in Acro-chordus granulatus (Schneider) (Reptilia Serpentes) with specialconsideration of the oticondashoccipital complex Journal of Morphol-ogy 249252ndash266

Russell D 1967 Systematics and morphology of American mosasaursPeabody Museum of Natural History Yale University Bulletin 231ndash241

Scanlon J D and M S Y Lee 2000 The Pleistocene serpent Won-ambi and the early evolution of snakes Nature 403416ndash420

Smith-Woodward A 1901 On some extinct reptiles from Patagonia ofthe genera Miolania Dinilysia and Genyodectes Proceedings ofthe Zoological Society London 1901169ndash184

Tchernov E M J Polcyn and L L Jacobs 2001 Application ofcomputed tomography to Pachyrhachis problematicus a snakewith legs from the Cretaceous of Israel Journal of Vertebrate Pa-leontology 21(3 suppl)107A

mdashmdashmdash O Rieppel H Zaher M J Polcyn and L L Jacobs 2000 Afossil snake with limbs Science 2872010ndash2012

Wever E G 1978 The Reptile Ear Its Structure and Function Prince-ton University Press Princeton 1024 pp

Zaher H and O Rieppel 1999 The phylogenetic relationships of Pa-chyrhachis problematicus and the evolution of limblessness insnakes (Lepidosauria Squamata) Comptes Rendus de lrsquoAcademiedes Sciences Paris 329831ndash837

Received 21 March 2002 accepted 17 June 2002