In the study of animal signals, spiders have emerged as aclassic example of signalling using substrate-propagatedvibrations (Barth, 1998). The vibrations propagated throughthe delicate webs of orb-weaving spiders are clear examples ofsignalling through vibrations (Barth, 1998; Finck, 1981;Frohlich and Buskirk, 1982; Klarner and Barth, 1982; Landolfaand Barth, 1996; Masters, 1984; Masters and Markl, 1981;Vollrath, 1979), but the majority of spiders may also usesubstrate-propagated vibrations in such varied substratesas water, soil, leaf litter or plants (Barth, 1985, 1998,2002; Bleckmann and Barth, 1984; Bristowe, 1929;Fernandezmontraveta and Schmitt, 1994; Rovner, 1968;Stratton and Uetz, 1983; Uetz and Stratton, 1982). Three typesof substrate-borne vibration-production mechanisms have beendescribed in spiders: percussion, stridulation and vibration(tremulation; Uetz and Stratton, 1982). Percussion is producedby the drumming of body parts against the substrate and hasbeen described in a variety of species (Dierkes and Barth,1995; Stratton, 1983; Uetz and Stratton, 1982). Stridulationoccurs by the rubbing of two rigid body structures relative toeach other (Dumortier, 1963) and seems to occur commonly inspiders (Legendre, 1963), particularly in wolf spiders (Family:
Lycosidae; Rovner, 1975; Stratton and Uetz, 1983; Uetz andStratton, 1982). Tremulation (Morris, 1980) is the third methodof substrate-borne vibration production found in spiders(Barth, 2002; Dierkes and Barth, 1995; Rovner, 1980; Uetz andStratton, 1982) and occurs by the oscillation of body parts,without a frequency multiplier (i.e. stridulation), coupled to thesubstratum, usually by adhesive hairs on the tips of one or moreof the legs. All of these mechanisms can be used to producesubstrate-borne (seismic) signals (Aicher et al., 1983; Aicherand Tautz, 1990; Narins, 1990).
Jumping spiders (Family: Salticidae) are unique amongspiders in that they are visual ‘specialists’, having two large,prominent frontal eyes that are specialized for high spatialresolution, as befits their predatory habits as stalker-hunters(Forster, 1982a; Land, 1985). Not surprisingly, vision alsoplays a prominent role in their signalling behaviour. Males,unlike females, have evolved conspicuously ornamented andcoloured appendages that they wave like semaphores duringcourtship, producing stereotyped, species-specific visualdisplays that unfold over periods of seconds to minutes(Crane, 1949; Forster, 1982b; Jackson, 1982). These displaysfunction in species isolation, species recognition and female
Visual displays in jumping spiders have long beenknown to be among the most elaborate animalcommunication behaviours. We now show that onespecies, Habronattus dossenus, also exhibits anunprecedented complexity of signalling behaviour in thevibratory (seismic) modality. We videotaped courtshipbehaviour and used laser vibrometry to record seismicsignals and observed that each prominent visual signal isaccompanied by a subsequent seismic component. Threebroad categories of seismic signals were observed(‘thumps’, ‘scrapes’ and ‘buzzes’). To furthercharacterize these signals we used synchronous high-speedvideo and laser vibrometry and observed that only oneseismic signal component was produced concurrently withvisual signals. We examined the mechanisms by whichseismic signals are produced through a series of signal
ablation experiments. Preventing abdominal movementseffectively ‘silenced’ seismic signals but did not affect anyvisual component of courtship behaviour. Preventingdirect abdominal contact with the cephalothorax, whilestill allowing abdominal movement, only silenced thumpand scrape signals but not buzz signals. Therefore,although there is a precise temporal coordination of visualand seismic signals, this is not due to a commonproduction mechanism. Seismic signals are producedindependently of visual signals, and at least threeindependent mechanisms are used to produce individualseismic signal components.
Seismic signals in a courting male jumping spider (Araneae: Salticidae)
Damian O. Elias1,*, Andrew C. Mason2, Wayne P. Maddison3 and Ronald R. Hoy11Department of Neurobiology and Behavior, Cornell University, Seeley G. Mudd Hall, Ithaca, NY 14853, USA,
2Division of Life Sciences, University of Toronto at Scarborough, 1265 Military Trail, Toronto, Ontario, Canada M161A4and 3Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ 85721, USA
choice (Clark and Morjan, 2001; Clark and Uetz, 1993;Jackson, 1982) and are specific enough to be useful astaxonomic characters (Richman, 1982). These displays aretextbook examples of visual communication (Bradbury andVehrencamp, 1998). While visual signals are well established,seismic signal production by stridulation (Edwards, 1981;Gwynne and Dadour, 1985; Maddison and Stratton, 1988),percussion (Noordam, 2002) and tremulation (Jackson, 1977,1982) has been proposed in a few species of jumping spiders.
Within the jumping spiders, members of the genusHabronattus are known for extraordinary diversity –especially of the complex, colourful ornaments used in theirmultifaceted visual displays (Griswold, 1987; Maddison andMcMahon, 2000; Peckham and Peckham, 1889, 1890). Over100 species have been described, with most of them occurringin North America, especially in arid regions of the southwest.Among these species, many exhibit striking morphologicaland geographical variation (Maddison and McMahon, 2000;Masta, 2000; Masta and Maddison, 2002). We focused onone particular species that has multiple, complex visualornaments: Habronattus dossenus. We recorded malecourtship behaviour in H. dossenusby using video and laservibrometry and found that the complex visual displays ofsignalling males represent only one component of anextremely elaborate multi-modal display. Male H. dossenussignal to prospective mates using a repertoire of seismicsignals coordinated with specific visual signals. In order toinvestigate these phenomena, we (1) characterized seismicand visual signals in detail using synchronous high-speedvideo and laser vibrometry and (2) examined possible seismicsignal production mechanisms by performing severalexperiments where we attempted to manipulate seismicsignals. We manipulated abdominal (opisthosoma)movements and contact with the cephalothorax (prosoma)because previous experiments in another Habronattus speciessuggested that seismic signal production originated there(Maddison and Stratton, 1988).
Materials and methodsCourtship behaviour of H. dossenus
Spiders
Male and female Habronattus dossenusGriswold werecollected in the field between July and September in 2000 and2001 from the Atascosa Mountains, Coronado NationalPark, southwestern Arizona (Santa Cruz County). Animalswere collected predominantly on leaf litter, rocks or sand.Animals were housed individually and kept in the lab on a12·h:12·h light:dark cycle. Weekly, spiders were fed a diet offruit flies (Drosophila melanogaster) and crickets (Achetadomesticus).
Recording procedures
We anaesthetized female H. dossenuswith CO2 and tetheredthem to a wire with low melting point wax. We held femalesin place with a micromanipulator on a substrate of stretched
nylon fabric (25·cm×30·cm). Males were then dropped ontothis substrate 15·cm from the female and allowed to courtfreely. Females were rotated to face the male until he orientedto her; recordings began when males approached females. Werecorded substrate vibrations produced during courtship usinga laser Doppler vibrometer (LDV; Polytec OFV 3001controller, OFV 511 sensor head; Waldbronn, Germany)(Michelsen et al., 1982). Laser Doppler vibrometry is a non-contact method of recording vibrations that measures thevelocity of a moving surface by detecting the Doppler shift ofa reflected laser beam. Pieces of reflective tape (approximately1·mm2) were attached to the underside of the courtshipsubstrate 2·mm from the female to serve as measurementpoints for the LDV. The LDV signal was synchronized withtwo concurrent methods of video recording: (1) the LDV signalwas recorded on the audio track during standard video tapingof courtship behaviour (Navitar Zoom 7000 lens; PanasonicGP-KR222; Sony DVCAM DSR-20 digital VCR; 44.1·kHzaudio sampling rate) or (2) the LDV signal was digitized (PCI-6023E; National Instruments, Austin, TX, USA; 10·kHzsampling rate) simultaneously with the capture of digitalhigh-speed video (500·frames·s–1; PCI 1000; RedLakeMotionscope, San Diego, CA, USA; Nieh and Tautz, 2000),using Midas software (Xcitex, Cambridge, MA USA). Allrecordings were made on a vibration-isolated table. In somerecordings, we also captured air-borne sound on a secondchannel using a probe microphone (B&K Type 4182, B&KNexus amplifier; Nærum, Denmark).
Sound and video analysis
Complete courtships of 20 different males were recorded.The same tethered female was used for all recordings.Examples were selected for detailed analysis. Bodymovements were measured frame-by-frame from digital high-speed video using Midas software. We calibrated absolutedistances by photographing a 1·mm2 grid before eachrecording. Power spectra of vibratory signals were calculatedusing Matlab software (The Mathworks, Natick, MA, USA).Spectrograms were made using Canary (Cornell University,Lab of Ornithology).
Seismic signal production mechanisms of H. dossenus
Experimental manipulations
For the signal manipulation experiments, the arena substratefloor for courtship was a sheet of graph paper attached toa square cardboard frame (60·cm×45·cm). Females weretethered as above, and the male’s seismic signals recorded usinga piezo-electric sensor placed directly underneath the tetheredfemale. We calibrated the response of the piezo-electric sensorusing a vibration source (B&K Type 4810 Mini-shaker) andLDV (OFV 3001 controller, OFV 511 sensor head). Althoughlow-frequency responses (<150·Hz) were relatively attenuatedby the piezoelectric sensor, the male’s signals were notsignificantly altered (data not shown). All experiments wereconducted in a sound-attenuated chamber. Seismic signals wereamplified (Nikko NA790) and recorded on the audio track of a
D. O. Elias and others
4031Seismic signalling in jumping spiders
video recording as above (44.1·kHz audio sampling rate). Allrecordings were also videotaped (Navitar Zoom 7000 lens;Panasonic GP-KR222; Sony DVCAM DSR-20 digital VCR).Recordings of signals were made from each male prior toexperimental manipulation. Classical spider anatomy hasrecognized two body segments in spiders: the prosoma andopisthosoma (Barth, 2002; Foelix, 1996). We use the alternativenomenclature, cephalothorax (prosoma) and abdomen(opisthosoma) to describe the spider’s body segments(Maddison and Stratton, 1988). We manipulated males by (1)preventing abdominal movements by attaching the abdomento the cephalothorax using wax (Kerr Sticky Wax; CencoScientific, Chicago, IL, USA; Fig.·3) and (2) preventing contactbetween the cephalothorax and abdomen by attaching a smallpiece of aluminium foil to the cephalothorax with wax; thisformed a flap that could be inserted at the junction between theabdomen and the cephalothorax (Fig.·5). To ensure that thesetreatments did not affect normal locomotory activities, wewaited two days following these manipulations and observedwhether or not the spiders were able to successfully captureprey. Both manipulations were reversible. Two days followingreversal by removing the wax or the foil flap, we recordedcourtship signals again. We used only males that were able tocapture prey during both intervals.
Power spectra analysis
Within a treatment set (control, experimental treatment,recovery) from an individual animal, individual signals (seebelow) were identified using videotaped data, and a randomselection of each seismic signal type acquired. The powerspectra of the noise floor, acquired before the start of everyrecording, was subtracted using Matlab software. Powerspectra of different signals were then calculated and averagedusing Matlab. This shows how, within an individual, the entirepower spectrum of a signal changes according to experimentaltreatment.
Statistical analysis
For each signal, peak intensities were recorded. For thumps,peak intensities below and above 500·Hz were recorded. Forscrapes, the peak intensity was recorded. For buzzes, theintensities of the first three harmonics were recorded. Withintreatment sets for each individual, intensities were normalizedto the highest intensity produced for all of the signalcomponents. Normalized intensities were then averaged andthe relative dB difference between the treatments calculated.The normalized intensities for different individuals were thenpooled into their treatment categories and averaged.Differences between treatments were tested for significance(P<0.05) using a repeated-measures analysis of variance(ANOVA) procedure and a post-hoc Tukey test withBonferonni corrections.
Scanning electron microscopy (SEM)
Specimens were fixed, dried and gold coated and thenviewed with a Philips SEM 505 microscope.
ResultsCourtship behaviour of H. dossenus
We divided courtship into four distinct phases based onvideo data (Fig.·1). Behaviourally, phase 1 consists of sidlingmovements in which the male approaches in a typical salticid‘zigzag’ visual display (Forster, 1982b). During this approach,the male waves his forelegs and spreads his pedipalps in astereotyped fashion. Phase 2 occurs when the male comes towithin approximately one body length (5–8·mm) of the femaleand produces rapid bouts of visible ‘downbeat’ gestures as hesettles into a typical courtship posture (Fig.·2i). Phase 3consists of multiple bouts of prolonged signalling. In phase 4,the male attempts to mount the female. Seismic displays occuronly in phases 2 to 4 (Fig.·1). Phase 2 is associated with a rapidbout of thumps (see below). Phase 3 consists of multiple boutsof signalling (thumps, buzzes and scrapes; see below). In phase4, the male accelerates the rate of signals, combiningpreviously separate signals (thumps and buzzes; see below). Atleast three signal types (thumps, scrapes and buzzes;Fig.·2iii,iv) were evident in all complete courtships, eachassociated with characteristic stereotyped body postures andunique foreleg movements (Fig.·2i,ii), abdominal movements(Fig.·2i), temporal characteristics (Fig.·2iii) and power spectra(Fig.·2iv). Seismic and visual signal components were onlyproduced during male and female interactions and never in anyother context. Analysis of video recordings showed thatseismic signals coincide with stereotyped movements of theabdomen and forelegs and both define and account for the threesignals described below (Fig.·2).
Thumps
Thumps (Fig.·2A) occur at the beginning of a sequence ofseismic signals. They can precede a sequence of scrape groupsor buzzes in phase 3 of courtship (Fig.·2A; see below) or occursimultaneously with buzzes in phase 4 (Fig.·1). The front legsand abdomen both produce the thump (Fig.·2A). First, theforelegs are raised high above the body and are then rapidlyslapped down onto the substrate (1–2 in Fig.·2A), producing apercussive impulse (2 in Fig.·2A). This percussive componentwas the only display that produced a detectable air-bornecomponent (data not shown). Approximately 8·ms later, theforelegs return to a nearly vertical position (2–3 in Fig.·2A)and the abdomen is pulled back and released (4–5 in Fig.·2A),causing it to ‘ring’ at a frequency of 58.3±7.5·Hz (mean ±S.D.,N= 5; Fig.·2Ai). This movement produces a brief, high-intensity broadband signal (Fig.·2Aiv). Movements of theforelegs and abdomen are highly coordinated, with delays of86.1±32.0·ms (N=27) for lone thumps and 46.0±8.0·ms (N=30)for thumps preceding buzzes. Both of these categories ofthumps also differ in duration and envelope shape (data notshown). Thumps consist of unique foreleg movements(Fig.·2Aii) and two seismic components: a percussivecomponent caused by the front legs contacting the substrateand a more-intense component caused by the oscillation of theabdomen (Fig.·2Aiv).
4032
Scrapes Scrapes (Fig.·2B) are emitted in groups lasting 5.3±1.1·s
(N=10) (Sc G in Fig.·1C). Within these groups, scrapes occurat a frequency of 5.7±1.2·Hz (N=15 scrape groups; Fig.·2Bi).One to four scrape groups occur between thumps and theseoccur only in phase 3 of courtship (Fig.·1). Individual scrapes(Sc in Fig.·1C) are associated with movements of the forelegsand abdomen (Fig.·2Bi). An up-and-down movement of theforeleg tips (2–3 in Fig.·2B) is followed by a dorso-ventraloscillation of the abdomen (1–2 in Fig.·2B). This ‘rockingmotion’ produces an underlying low-frequency oscillation(5.7·Hz) that is evident in the oscillogram (Fig.·2Biii).Abdominal and foreleg movements are highly coordinated, with
delays of 32.3±7.0·ms (N=409). In adjacent scrape groups, theforelegs alternate coming together and moving apart laterally.Two types of movements can occur between scrape groups: (1)when scrape groups follow thumps, the 3rd legs are re-positioned against the body as the male moves forwardincrementally, and (2) when scrape groups precede a thump, thepedipalps are moved rapidly up and down prior to the thump.Individual scrape seismic signals are produced only duringabdominal movements (Fig.·2B) and not during characteristicforeleg movements (Fig.·2Bii). Within groups, individualseismic scrapes are short, broadband signals (Fig.·2Biv). Thefrequency of abdominal movement is much lower than thefrequency of vibrational signal produced (Fig.·2Bi,iv).
Buzzes
Buzzes (Fig.·2C) occur alone in phase 3 ofcourtship or simultaneously with thumps inphase 4 (Fig.·1). Buzzes in phase 3 are alwayspreceded by 2–5 thumps. The number ofthumps occurring increases linearly ascourtship progresses (Fig.·1). Both abdominaland leg movements accompany the signal. Thefront legs come down in a slow continuousmovement (1–2 in Fig.·2C), while theabdomen produces a sustained, rapid, low-amplitude oscillation at a frequency of 65.0·Hz(Fig.·2C). Abdominal movements areprecisely synchronized with the vibratorysignal, while distinctive foreleg movements(Fig.·2Cii) occur at variable delays(180±644·ms, N=14; Fig.·2C). Buzz seismicsignals are long in duration, with afundamental frequency of 65.0±2.9·Hz(N=12) plus higher harmonics (Fig.·2Civ).Frequencies of seismic buzzes are temperaturedependent (data not shown). Abdominaloscillations are at the same 65·Hz frequency asthe fundamental frequency of the buzz seismicsignal (Fig.·2C).
Seismic signal production mechanisms of H.dossenus
Experimental manipulations
Abdominal movement. Analysis of high-speed videos, along with observationssuggesting that abdominal movements are notvisible to a female while the male is courting,suggests that most seismic signals areproduced by abdominal movements and not bymovements of the legs. To investigate whetherseismic signals are produced by any of theobserved body movements, we performed aseries of experiments where we tried toeliminate signals. We did this by immobilizingthe abdomens of males by fixing them withwax to the cephalothorax (Fig.·3). This
Fig.·1. Seismic signals of courting male jumping spiders. (A) Sonogram of a seismicsignal. (B) Oscillogram of seismic signals. Courtship can be divided into four distinctphases, with seismic signals occurring in phases 2–4. (C) Detail of oscillogram markedby the box in B. All three types of seismic signals can be observed: thumps [Th (red)],buzzes [Bz (green)] and scrapes [Sc (blue)]. Individual scrapes occur in groupsconsisting of multiple repeated scrapes [Sc G (yellow)]. Recordings made using a laserDoppler vibrometer.
4033Seismic signalling in jumping spiders
treatment was fully reversible. Males were recorded prior totreatment, then with abdomen immobilized and finally afterremoval of the wax. We could readily identify the occurrenceof each signal type by the stereotypic leg movements andpostures characteristic of each signal from videotapes(Fig.·2ii). Only the abdominal and not the weak percussivecomponent of the thump was analyzed (Fig.·2A). All threeseismic signals were greatly attenuated when the abdomen wasimmobilized (Figs·3,·4). All frequencies were attenuated inall signal types (Fig.·3). Experimental treatments weresignificantly different (P<0.001) from both control andrecovery treatments (Fig.·4). All signals recovered followingremoval of the wax, and no significant differences wereobserved between the control and recovery treatments (Fig.·4).Thus, abdominal movements are necessary for seismicsignalling.
Abdomen–cephalothorax contact. Observations usingsynchronous high-speed video and vibrational recordingsrevealed that the power spectrum of a buzz exactly matchedthe oscillation frequency of the abdomen, while the powerspectra of thumps and scrapes included much higherfrequencies than the oscillation frequency of the abdomen.This hinted that buzz, scrape and thump signals are producedby different mechanisms. Hence, in a second set ofexperiments, we prevented direct contact between thecephalothorax and abdomen but did not prevent abdominalmovements (Fig.·5). We prevented abdomen–cephalothoraxcontact by placing a small barrier of aluminium foil betweenthe cephalothorax and abdomen. Recovery treatmentsconsisted of removing the barrier (Fig.·5). Buzzes wereunaffected at all frequencies (Figs·5A,·6C); no significantdifferences were observed between the control, experimental
Fig.·2. Types of seismic signals. Top panels (i) show body positions, with numbers (1–5) illustrating movements of the forelegs and abdomen.Middle panels show (ii) the position of one of the forelegs (mm above the substrate) and (iii) the oscillograms of the seismic signals. Bottompanels (iv) show the frequency characteristics of the seismic signals. Panels ii–iv are shown in the same time scale, with numbers (1–5)corresponding to the body movements illustrated in panel i. (A) Thump signal. Front legs come down (1–2), contact the substrate and quicklymove back up (2–3). Shortly afterwards the abdomen is pulled back and released, and the abdomen ‘rings’ at 58.3·Hz (4–5). Thumps arebroadband signals with peak frequencies at 203·Hz and 1203·Hz. Production of signal corresponds with the percussive contact of the front legsagainst the substrate (1–2) and movements of the abdomen (4–5). (B) Scrape signal. Abdomen moves up (1–2) and shortly afterwards the frontlegs come down (2–3). Scrapes occur in groups with a frequency of 5.7·Hz. Scrapes are broadband signals with peak frequencies at 230·Hz and550·Hz. Production of seismic signal corresponds to movements of the abdomen. (C) Buzz signal. Front legs come down (1–2) as the abdomenoscillates at 65·Hz (1–2). This signal has a fundamental frequency at 65·Hz with several harmonic frequencies (130·Hz, 195·Hz and 260·Hz).Production of seismic signal corresponds with movements of the front legs and abdomen.
3 3 3
1 1 12 2 2
2
31
4
5
Thump
i
A B C
ii
iii
2
3 2
1
Scrape
65 Hz5.7 Hz
58.3 Hz
1
21–2
Buzz
1 2 3
5
5
2
3 3
12 2
3
1
2
1 2
Time (s)
0 31 2
Time (s)
0 0.1
5.6
5.4
5.2
5
4.2
4
3.8
3.6
0.2 0.3 0.4 0.5
32 32 2 31 1 1
21
4
31
4 1 12
Vib
ratio
nam
plitu
deD
ista
nce
(mm
off
subs
trat
e)
iv
Freq
uenc
y(H
z)
Time (s)
1000
1500
500
00 0.05 0.1 0.15 0.2 0.25
45
2
1
3
3
2
1
0
4034 D. O. Elias and others
Buzz
i ii iii
A
ScrapeB
Pow
er
ThumpC
Frequency (Hz)
Control Experimental treatment Recovery
0 500400300200100
N=7 N=7
600
10–8
10–6
10–4
10–2
0 500400300200100 600
10–8
10–6
10–4
10–2
0 500400300200100 600
10–8
10–6
10–4
10–2
0 50004000300020001000
10–9
10–7
10–6
10–5
10–8
0 50004000300020001000
10–9
10–7
10–6
10–5
10–8
0 50004000300020001000
10–9
10–7
10–6
10–5
10–8
0 50004000300020001000
10–8
10–6
10–4
10–8
10–6
10–4
0 40005000300020001000
10–8
10–6
10–4
0 40005000300020001000
N=5
N=23 N=23 N=23
N=19 N=19 N=19
Fig.·3. Effects of male abdominal immobilization on power spectra of different seismic signals. (A) Buzz signal; (B) scrape signal; (C) thumpsignal. Panels i–iii represent mean power spectra for one individual during the control, experimental and recovery treatments, respectively.Experimental treatment consisted of waxing the cephalothorax to the abdomen, rendering body segments immovable relative to each other.Recovery treatment consisted of removing the wax from the animal.
Fig.·4. Effects of male abdominalimmobilization on (A) thump,(B) scrape and (C) buzz seismicsignals. Within individuals, peakintensities were normalized to themaximum intensity produced forall of the signal components.Normalized intensities were thenaveraged, and the relative dBdifference between the treatmentscalculated. Graphs show relativedB difference between thetreatments (control, experimentaltreatment and recovery) of allthe individuals tested ±S.D.(N=5). Experimental treatmentsattenuated peak frequencies of all signals significantly (**P<0.001; Tukey post-hoc test with Bonferonni corrections). No significantdifferences were observed between control and recovery treatments.
–34
–26
–18
–10
–2
6
A B C
ControlTreatmentRecovery
Low(<500
Hz)
High(>500
Hz)
Rel
ativ
e dB
** ****
**
Frequency
Thump
–28
–20
–12
–4
4
Rel
ativ
e dB
Peak frequency
****
Scrape
–45
–35
–25
–15
–5
5
1st 2ndHarmonic
3rd
** **
Buzz
Rel
ativ
e dB
** ****
**
4035Seismic signalling in jumping spiders
Buzz
i ii iii
A
ScrapeB
Pow
er
ThumpC
Frequency (Hz)
Control Experimental treatment Recovery
0 500400300200100
N=6 N=6
600
10–6
10–4
10–2
0 500400300200100 600
10–6
10–4
10–6
10–4
10–2
0 500400300200100 600
10–2
0 50004000300020001000
10–5
0 50004000300020001000
10–5
0 50004000300020001000
10–7
10–6
10–5
10–8
10–7
10–6
10–8
10–7
10–6
10–8
0 50004000300020001000
10–8
10–6
10–4
10–8
10–6
10–4
0 40005000300020001000
10–8
10–6
10–4
0 40005000300020001000
N=10
N=23 N=20 N=23
N=19 N=19 N=16
Fig.·5. Effects of preventing male abdominal and cephalothorax contact on the power spectra of different seismic signals. (A) Buzz signal; (B)scrape signal; (C) thump signal. Panels i–iii represent mean power spectra for one individual during the control, experimental and recoverytreatments, respectively. Experimental treatment consisted of waxing a piece of flexible foil to the cephalothorax and placing one end of the foilbetween the cephalothorax and abdomen. Recovery treatment consisted of removing the foil from between the cephalothorax and abdomen.
Fig.·6. Effects of preventingmale abdominal andcephalothorax contact on (A)thump, (B) scrape and (C)buzz seismic signals. Withinindividuals, peak intensitieswere normalized to themaximum intensity producedfor all of the signal components.Normalized intensities werethen averaged, and the relativedB difference between thetreatments calculated. Graphsshow relative dB differencebetween the treatments (control,experimental treatment andrecovery) of all the individuals tested ±S.D. (N=5). Experimental treatments attenuated peak frequencies of scrape and high-frequency(>500·Hz) ranges of thumps significantly (*P<0.05; Tukey post-hoctest with Bonferonni corrections). No significant differences were observedfor buzz and low (<500·Hz)-frequency ranges of thumps. No significant differences were observed between control and recovery treatments.
–11
–8
–5
–2
1
4
Rel
ativ
e dB
1st 2ndHarmonic
3rd
Buzz
–28
–21
–14
–7
0
7
Rel
ativ
e dB
Peak frequency
**
Scrape
–32
–24
–16
–8
0
8
A B C
TreatmentRecovery
Control
Low(<500
Hz)
High(>500
Hz)
Rel
ativ
e dB
*
Frequency
*
Thump
4036
and recovery treatments (Fig.·6C). Both scrapes and thumps,however, were affected. Scrapes were attenuated significantlyat all frequencies (Figs·5B,·6B). For thumps, low-frequencycomponents (<500·Hz) were unaffected but high-frequencycomponents of the thump (>500·Hz) were attenuated(Figs·5C,·6A). Experimental treatments for the scrape andhigh-frequency components of the thump were significantlydifferent (P<0.05) from both control and recovery treatments(Fig.·6). Control and recovery treatments were similar forall components (Fig.·6). Thus, including the percussivecomponent of thumps, at least three separate mechanisms areused in the production of vibrational signals. Buzz signals areproduced by abdominal oscillations and do not require contactbetween the abdomen and cephalothorax. Scrape and thumpsignals, on the other hand, require abdomen–cephalothoraxcontact to produce the high frequencies evident in both of thesesignals.
Scanning electron microscopy
The observation that high-frequency signal componentsrequire direct contact between body parts that move relative toeach other suggests a stridulatory mechanism (Dumortier,1963). Therefore, we examined, using SEM, thecephalothorax–abdomen junction of both male and female H.dossenusfor evidence of a stridulatory apparatus, as observedin males of another Habronattus species (Maddison and
Stratton, 1988). Female H. dossenusdo not produce seismicsignals in any context. SEMs revealed the presence of a file onthe male cephalothorax (Fig.·7Bi) but not on the female(Fig.·7Ai). In the apposing abdominal areas, we noted thepresence of hardened sclerotized scrapers on the male(Fig.·7Bii) but not on the female (Fig.·7Aii). Thus, scrape andthump signals appear to be produced by stridulation.
DiscussionOur results show that male H. dossenus use seismic signals
together with their visual displays and that male H. dossenuscourtship signals consist of complex visual signals co-occurring with multiple seismic signals. Based on high-speedvideo and synchronous laser vibrometer recordings, seismicsignals correspond to movements of the male’s abdomen butnot of his forelegs (with the exception of the initial percussivecomponent of thumps). Furthermore, preventing abdominalmovements by fixing the abdomen relative to thecephalothorax ‘silenced’ males but did not affect visual orpercussive display components. Hence, visual and seismicsignals are produced by anatomically different neuromuscularmechanisms, visual signals by muscles controlling forelegmovement and vibratory signals by muscles controllingabdominal movement, yet both signals are coordinated withdelays of 30–60·ms for scrape and thumps and 300·ms for
D. O. Elias and others
F
SS
SS
i Cephalothorax ii Abdomen
0.1 mm
Male
B
AFemale
Fig.·7. Scanning electron micrograph (SEM) of cephalothorax and abdomen junction on (A) female and (B) male H. dossenus. (i) SEM of theposterior end of the head; (ii) SEM of the anterior end of the abdomen. F represents the ridged file found on male H. dossenus. S shows thelocation of the scrapers on the male.
4037Seismic signalling in jumping spiders
buzzes. Abdominal movements exactly match the frequencycharacteristics of the buzz signal but not the thump or scrapesignal. In addition, preventing contact between the abdomenand the cephalothorax attenuated thump and scrape signals butnot buzz signals. SEMs of the cephalothorax–abdomenjunction revealed the presence of a scraper and file; thus,scrapes and thumps are produced through stridulation whilebuzzes are not.
Three different mechanisms are responsible for the differentsignals: (1) the first thump component is produced frompercussion with the forelegs and the ground, (2) scrapes andthe second thump component are produced from abdominalmovements coupled to a frequency multiplier (stridulation) and(3) buzzes are produced from abdominal oscillations alone(tremulation). Selective elimination of only the highfrequencies of thump signals suggests that both vibratorymechanisms (stridulation and tremulation) contribute tothumps or possibly that low frequencies in thumps areproduced using a different area on the scraper, one wherecontact was not prevented. The entire diversity of substrate-borne vibration-production mechanisms described to date inspiders (Uetz and Stratton, 1982) is seen here in one species:H. dossenus. To our knowledge, no other spider describedexhibits such complexity in seismic signal production. This issurprising since it occurs in a family in which signalling isthought to be predominantly visual (Foelix, 1996). This raisesthe question of why H. dossenushas evolved multiple seismicsignals in addition to its repertoire of visual signals.
Two major ‘quality-based’ hypotheses have been proposedfor the evolution of multiple signals: ‘backup signals’ and‘multiple messages’ (Johnstone, 1996; Moller andPomiankowski, 1993). The backup signals hypothesis statesthat different signals provide the same information about asender but allow for a more accurate assessment of condition,while the multiple messages hypothesis states that differentsignals code for different aspects of a senders condition. Thebackup hypothesis, in this context, would predict that visualand seismic signals are alternative media for the same signalinformation and that seismic signals may be most importantwhen visual signals are obscured. This seems unlikely forseveral reasons. H. dossenuscourtship only occurs diurnally.Visual courtship starts at ranges up to 60·mm away whileseismic courtship signals only occur at close ranges (5–8·mm).The start of courtship appears to be visually mediated sincemales orient and court to tethered females in the absence ofany chemical cues produced, for example, by the female’s dragline. Sometimes, however, males will display when the femaleis looking in the opposite direction. Regardless, courting malesare usually in the female’s line of sight and in close proximitywhen seismic signals are produced. This still leaves thequestion of whether the three different seismic signals areacting as ‘backups’ to each other. H. dossenuscan be collectedon various substrates; leaf litter, sandy soil or rocks. Each ofthe different substrates has very different transmissionproperties (D. O. Elias, R. R. Hoy and A. C. Mason, manuscriptin preparation) and it is possible that some signals propagate
better in some substrates than others. The difference at shortdistances is minimal however. Also, the most commonsubstrate (leaf litter) transmits all signals equally well. Again,because all signals are produced at very close distances, wheresignal attenuation is presumably negligible, it seems unlikelythat the three signals are redundant backups. Anotherpossibility is that the different seismic signal productionmechanisms may act to backup one another. This is unlikelydue to the large temporal and spectral differences between thesignals.
A better alternative is that seismic signals are used asmultiple messages for sender condition. Male H. dossenushavemultiple visual ornaments. Males, but not females, arestrikingly ornamented, especially the body parts that are usedin courtship. The forelegs, for example, are bright green witha dark brown border and a fringe of white hair, while the tipsof the legs are a deep black. The pedipalps, third pair of legsand face are also ornamented (Griswold, 1987). One problemthat may be encountered by having multiple signals in a singlemodality is the amount of information that can be effectivelydetected and discriminated (Rowe, 1999). Within a discretesignal modality, habituation, adaptation and transductionmechanisms in sensory neurons, as well as memorycapabilities of receivers, may set limits to signals that animalsare able to effectively detect and process. Complex signalswith many different characteristics in a single modality, forexample, are often perceived as one unified stimulus (Honeyand Hall, 1989; Rowe, 1999), while information transmitted inmultiple modalities is not (Hillis et al., 2002). The evolutionof seismic signals could therefore be a way to add multiplemessages when there is selection for multiple avenues ofinformation for females and the evolution of further signals inthe visual modality is limited by physiological or economicconstraints. The three different seismic signals could also beused to relay multiple messages. The occurrence of threedifferent seismic signal production mechanisms that involvedifferent motions and anatomical structures suggests thepossibility that each different signal could relay very differentinformation about the male’s condition.
Alternatives to these two quality-based hypotheses havebeen proposed in models of the evolution of multiple sexualpreferences and ornaments (Iwasa and Pomiankowski, 1994;Pomiankowski and Iwasa, 1993, 1998). These models are notnecessarily based on mate quality assessment but are insteadbased on Fisherian ‘runaway selection’ (Fischer, 1930) andtheir interplay with other Fisherian and handicap traits (Zahavi,1975). In these models, female preferences lead to theelaboration of male display traits, and multiple male ornamentsevolve in spite of the increased cost to males.
Regardless of the evolutionary process that has led to signalelaboration in this species, a further question is how theaddition of a second stimulus modality contributes to signalcontent and efficacy. Spiders in the Habronattusgroup areknown for the complexity of visual displays as well as theirvisual ornaments. Habronattus dossenusis no exception to thispattern. How then does H. dossenusincorporate two separate
4038
but precisely coordinated sets of complex signals? Onepossibility is that it is the coordination of visual and seismicsignals that relays information. Especially with thumpspreceding buzzes and scrape signals, the coordination of visualand seismic signals can be very precise and it is possible thatfemales are using this tight temporal coordination as a measureof male quality. Another possibility is that either vibratory orvisual signals carry information, and the tight coordination ofthe alternative modality directs attention to subsequent signals.In animal signals, signal components that precede focalinformative signals have often been shown to improve signalefficacy and efficiency by directing attention (Fleishman,1988; Rowe, 1999). Jumping spiders’ well-developed sense ofvision (Blest et al., 1981; Eakin and Brandenburger, 1971;Forster, 1982a,b; Jackson, 1982; Land, 1969, 1985) couldpossibly be a good mechanism to draw attention to seismicsignals that may be more difficult to detect than visual signals.Alternatively, experiments in humans have shown that whensound stimuli are matched with a corresponding visualstimulus, the perception of visual temporal rate is improved(Recanzone, 2003). A similar process may be at work in theseanimals, particularly in the short-duration thump and scrapesignals, where seismic signals could improve the detection ofrapidly occurring visual signals. Similar arguments have beenmade regarding the evolution of visual ornaments and visualmotion displays in jumping spiders (Peckham and Peckham,1889, 1890). If visual form and motion pathways areconsidered separately (Barth, 2002; Forster, 1982b; Strausfeldet al., 1993), then it is possible that visual ornaments can focusattention on motions or vice versa(Hasson, 1991), hence thetight correlation between ornaments and the body parts used indisplays (Peckham and Peckham, 1889, 1890). Attentionfocusing via the combinatorial possibilities created by thecombination of multiple modalities or components couldtherefore be a powerful force driving the evolution of complexsignals.
While much recent work has documented the occurrence ofmulti-modal signals in a variety of animals (Fusani et al., 1997;Hoelldobler, 1999; Hughes, 1996; McGurk and MacDonald,1976; Partan and Marler, 1999; Rowe and Guilford, 1999),including spiders (Hebets and Uetz, 1999; Scheffer et al., 1996;Uetz and Roberts, 2002), the complexity found in H. dossenusis impressive. Multiple visual ornaments and visual displaysexist together with a complexity of seismic signals that isunprecedented in spiders. H. dossenususes three independentmechanisms to produce three types of signals, which canfurther be divided into at least seven categories based on thepower spectra, envelope shape and temporal structure of thesignals (D. O. Elias, A. C. Mason, W. P. Maddison and R. R.Hoy, unpublished observations). Each of these seismic signalsis precisely coordinated with a unique visual display, and somevisual signals (i.e. pedipalp signals) have no correspondingseismic component. We feel that this signal complexityrepresents a good system in which to test competing models ofsignal evolution. Future studies will examine female responsesto manipulated and control males in an attempt to elucidate the
function of different aspects of the male’s complex, multi-modal multi-component courtship signals.
We would like to thank B. Land, A. Spence, B.Wyttenbach, C. Clark, N. VanderSal, E. Hebets, K. Pilz, M.Andrade, M. Hedin and members of the Hoy lab for helpfulcomments, suggestions and assistance. Illustrations providedby Margy Nelson. Funding was provided by NIH (to R.R.H.;N1DCR01 DC00103), NSERC (to A.C.M.; 238882 241419)and a HHMI Pre-Doctoral Fellowship (to D.O.E.).
ReferencesAicher, B., Markl, H., Masters, W. M. and Kirschenlohr, H. L. (1983).
Vibration transmission through the walking legs of the fiddler crab, Ucapugilator (Brachyura, Ocypodidae) as measured by laser Dopplervibrometry. J. Comp. Physiol.150, 483-491.
Aicher, B. and Tautz, J. (1990). Vibrational communication in the fiddlercrab, Uca pugilator. 1. Signal transmission through the substratum. J. Comp.Physiol. A166, 345-353.
Barth, F. G. (1985). Neuroethology of the spider vibration sense. InNeurobiology of Arachnids(ed. F. G. Barth), pp. 203-229. New York:Springer-Verlag.
Barth, F. G. (1998). The vibrational sense of spiders. In ComparativeHearing: Insects(ed. R.-R. Hoy, A. N. Popper and R.-R. Fay), pp. 228-278.New York: Springer.
Barth, F. G. (2002). A Spider’s World: Senses and Behavior. Berlin,Heidelberg, New York: Springer-Verlag.
Bleckmann, H. and Barth, F. G. (1984). Sensory ecology of a semi aquaticspider Dolomedes triton. 2. The release of predatory behavior by watersurface waves. Behav. Ecol. Sociobiol.14, 303-312.
Blest, A. D., Hardie, R. C., McIntyre, P. and Williams, D. S. (1981). Thespectral sensitivities of identified receptors and the function of retinal tieringin the principal eyes of a jumping spider. J. Comp. Physiol.145, 227-239.
Bradbury, J. W. and Vehrencamp, S. L. (1998). Principles of AnimalCommunication. Sunderland, MA: Sinauer Assoc. Inc.
Bristowe, W. S. (1929). The mating habits of spiders, with special referenceto the problems surrounding sex dimorphism. Proc. Zool. Soc. 21, 309-358.
Clark, D. L. and Morjan, C. L. (2001). Attracting female attention: theevolution of dimorphic courtship displays in the jumping spider Maeviainclemens(Araneae: Salticidae). Proc. R. Soc. Lond. Ser. B. Biol. Sci.268,2461-2465.
Clark, D. L. and Uetz, G. W. (1993). Signal efficacy and the evolution ofmale dimorphism in the jumping spider, Maevia inclemens. Proc. Natl.Acad. Sci. USA90, 11954-11957.
Crane, J. (1949). Comparative biology of salticid spiders at Rancho Grande,Venezuela. Part IV. An analysis of display. Zoologica34, 159-214.
Dierkes, S. and Barth, F. G. (1995). Mechanism of signal production in thevibratory communication of the wandering spider Cupiennius getazi(Arachnida, Araneae). J. Comp. Physio. A 176, 31-44.
Dumortier, B. (1963). Sound Emissions in Arthropoda. In Acoustic Behaviorof Animals(ed. R. G. Busnel), pp. 277-338. Amsterdam, London, NewYork: Elsevier Science.
Eakin, R. M. and Brandenburger, J. L. (1971). Fine structure of the eyesof jumping spiders. J. Ultrastruct. Res.37, 618-663.
Edwards, G. B. (1981). Sound production by courting males of Phidippusmystaceus(Araneae: Salticidae). Psyche88, 199-214.
Fernandezmontraveta, C. and Schmitt, A. (1994). Substrate-bornevibrations produced by male Lycosa tarentula fasciiventris(Araneae,Lycosidae) during courtship and agonistic interactions. Ethology97, 81-93.
Finck, A. (1981). The lyriform organ of the orb-weaving spider Araneoussericatus– vibration sensitivity is altered by bending the leg. J. Acoust. Soc.Am.70, 231-233.
Fischer, R. A. (1930). The Genetical Theory Of Natural Selection. Oxford:Clarendon Press.
Fleishman, L. J. (1988). Sensory influences on physical design of a visual-display. Anim. Behav.36, 1420-1424.
Foelix, R. F. (1996). The Biology of Spiders. New York: Oxford UniversityPress.
D. O. Elias and others
4039Seismic signalling in jumping spiders
Forster, L. (1982a). Vision and prey-catching strategies in jumping spiders.Am. Sci.70, 165-175.
Forster, L. (1982b). Visual communication in jumping spiders (Salticidae).In Spider Communication: Mechanisms and Ecological Significance(ed. P.N. Witt and J. S. Rovner), pp. 161-212. Princeton: Princeton UniversityPress.
Frohlich, C. and Buskirk, R. E. (1982). Transmission and attenuation ofvibration in orb spider webs. J. Theor. Biol.95, 13-36.
Fusani, L., Hutchison, R. E. and Hutchison, J. B. (1997). Vocal-posturalco-ordination of a sexually dimorphic display in a monomorphic species:the barbary dove. Behaviour134, 321-335.
Griswold, C. E. (1987). A revision of the jumping spider genus HabronattusF.O.P. Cambridge (Araneae: Salticidae) with phenetic and cladisticanalyses. Univ. California Publ. Entomol. 107, 1-344.
Gwynne, D. T. and Dadour, I. R. (1985). A new mechanism of soundproduction by courting male jumping spiders (Araneae: Salticidae, Saitismichalseni Simon). Zool. Soc. Lond.207, 35-42.
Hasson, O. (1991). Sexual displays as amplifiers – practical examples with anemphasis on feather decorations. Behav. Ecol.2, 189-197.
Hebets, E. A. and Uetz, G. W. (1999). Female responses to isolated signalsfrom multimodal male courtship displays in the wolf spider genus Scizocosa(Aranerae: Lycosidae). Anim. Behav.57, 865-872.
Hillis, J. M., Ernst, M. O., Banks, M. S. and Landy, M. S. (2002).Combining sensory information: mandatory fusion within, but not between,senses. Science298, 1627-1630.
Hoelldobler, B. (1999). Multimodal signals in ant communication. J. Comp.Physiol. A.184, 129-141.
Honey, R. C. and Hall, G. (1989). Attenuation of latent inhibition aftercompound pre- exposure: associative and perceptual explanations. Q. J. Exp.Psychol. 41B, 335-368.
Hughes, M. (1996). The function of concurrent signals: visual and chemicalcommunication in snapping shrimp. Anim. Behav.52, 247 257.
Iwasa, Y. and Pomiankowski, A. (1994). The evolution of mate preferencesfor multiple sexual ornaments. Evolution48, 853-867.
Jackson, R. R. (1977). Courtship versitility in the jumping spider, Phidippusjohnosni(Araneae: Salticidae). Anim. Behav.25, 953-957.
Jackson, R. R. (1982). The behavior of communicating in jumping spiders(Salticidae). In Spider Communication: Mechanisms and EcologicalSignificance(ed. P. N. Witt and J. S. Rovner), pp. 213-247. Princeton:Princeton University Press.
Johnstone, R. A. (1996). Multiple displays in animal communication: ‘backupsignals’ and ‘multiple messages’. Philos. Trans. R. Soc. Lond. Ser. B. Biol.Sci.351, 329-338.
Klarner, D. and Barth, F. G. (1982). Vibratory signals and prey capture inorb-weaving spiders (Zygiella x-notata, Nephila clavipes, Araneidae). J.Comp. Physiol.148, 445-455.
Land, M. F. (1969). Structure of retinae of principal eyes of jumping spiders(Salticidae – Dendryphantinae) in relation to visual optics. J. Exp. Biol. 51,443-470.
Land, M. F. (1985). The morphology and optics of spider eyes. InNeurobiology of Arachnids(ed. F. G. Barth). Berlin, Heidelberg, New York,Tokyo: Springer-Verlag.
Landolfa, M. A. and Barth, F. G. (1996). Vibrations in the orb web of thespider Nephila clavipes: cues for discrimination and orientation. J. Comp.Physiol. A179, 493-508.
Legendre, R. (1963). L’audition et l’e’mission de sons chez les araneides.Annal. Biol.2, 371-390.
Maddison, W. and McMahon, M. (2000). Divergence and reticulationamong montane populations of a jumping spider (Habronattus pugillisGriswold). Syst. Biol.49, 400-421.
Maddison, W. P. and Stratton, G. E. (1988). Sound production andassociated morphology in male jumping spiders of the Habronattus agilisspecies group (Araneae, Salticidae). J. Arachnol.16, 199-211.
Masta, S. E. (2000). Phylogeography of the jumping spider Habronattuspugillis (Araneae: Salticidae): recent vicariance of sky island populations?Evolution54, 1699-1711.
Masta, S. E. and Maddison, W. P. (2002). Sexual selection drivingdiversification in jumping spiders. Proc. Natl. Acad. Sci. USA99, 4442-4447.
Masters, W. M. (1984). Vibrations in the orbwebs of Nuctenea sclopetaria(Araneidae). 1. Transmission through the Web. Behav. Ecol. Sociobiol.15,207-215.
Masters, W. M. and Markl, H. (1981). Vibration signal transmission inspider orb webs. Science213, 363-365.
McGurk, H. and MacDonald, J. (1976). Hearing lips and seeing voices.Nature264, 746-748.
Michelsen, A., Fink, F., Gogala, M. and Traue, D. (1982). Plants astransmission channels for insect vibrational songs. Behav. Ecol. Sociobiol.11, 269-281.
Moller, A. P. and Pomiankowski, A. (1993). Why Have birds got multiplesexual ornaments. Behav. Ecol. Sociobiol.32, 167-176.
Morris, G. K. (1980). Calling display and mating-behavior of CopiphorarhinocerosPictet (Orthoptera, Tettigoniidae). Anim. Behav.28, 42-51.
Narins, P. M. (1990). Seismic communication in anuran amphibians.Bioscience40, 268-274.
Nieh, J. C. and Tautz, J. (2000). Behaviour-locked signal analysis revealsweak 200-300·Hz comb vibrations during the honeybee waggle dance. J.Exp. Biol.203, 1573-1579.
Noordam, A. P. (2002). Abdominal percussion and ventral scutum in maleEuophrys frontalis(Araneae: Salticidae).Entomol.e Berich. Amsterdam62,17-19.
Partan, S. and Marler, P. (1999). Communication goes multimodal. Science283, 1272-1273.
Peckham, G. W. and Peckham, E. G. (1889). Observations on sexualselection in spiders of the family Attidae. Occas. Pap. Wisc. Nat. Hist. Soc.1, 3-60.
Peckham, G. W. and Peckham, E. G. (1890). Additional observations onsexual selection in spiders of the family Attidae, with some remarks on Mr.Wallace’s theory of sexual ornamentation. Occas. Pap. Wisc. Nat. Hist. Soc.1, 117-151.
Pomiankowski, A. and Iwasa, Y. (1993). Evolution of multiple sexualpreferences by Fisher runaway process of sexual selection. Proc. R. Soc.Lond. Ser. B. Biol. Sci.253, 173-181.
Pomiankowski, A. and Iwasa, Y. (1998). Runaway ornament diversitycaused by Fisherian sexual selection. Proc. Natl. Acad. Sci. USA95, 5106-5111.
Recanzone, G. H. (2003). Auditory influences on visual temporal rateperception. J. Neurophysiol.89, 1078-1093.
Richman, D. B. (1982). Epigamic display in jumping spiders (Araneae,Salticidae) and its use in systematics. J. Arachnol.10, 47-67.
Rovner, J. S. (1968). An analysis of display in the Lycosa rabidaWalckenaer.Anim. Behav.16, 358-369.
Rovner, J. S. (1975). Sound production by nearctic wolf spiders – substratum-coupled stridulatory mechanism. Science190, 1309-1310.
Rovner, J. S. (1980). Vibration in Heteropoda venatoria(Sparassidae) – a 3rdmethod of sound production in spiders. J. Arachnol.8, 193-200.
Rowe, C. (1999). Receiver psychology and the evolution of multicomponentsignals. Anim. Behav.58, 921-931.
Rowe, C. and Guilford, T. (1999). The evolution of multimodal warningdisplays. Evol. Ecol.13, 655-671.
Scheffer, S. J., Uetz, G. W. and Stratton, G. E. (1996). Sexual selection,male morphology, and the efficacy of courtship signalling in two wolfspiders (Araneae: Lycosidae). Behav. Ecol. Sociobiol.38, 17-23.
Stratton, G. E. (1983). Comparison of courtship behaviors and interspecificcrosses in the Schizocosa ocreata species complex Araneae Lycosidae. Am.Zool. 23, 967.
Stratton, G. E. and Uetz, G. W. (1983). Communication via substrate-coupled stridulation and reproductive isolation in wolf spiders (Aranea:Lycosidae). Anim. Behav.31, 164-172.
Strausfeld, N. J., Weltzien, P. and Barth, F. G. (1993). Two visual systemsin one brain: neuropils serving the principle eyes of the spider Cupienniussalei. J. Comp. Neurol.328, 63-72.
Uetz, G. W. and Roberts, J. A. (2002). Multisensory cues and multimodalcommunication in spiders: Insights form video/audio playback studies.Brain Behav. Evol.59, 222-230.
Uetz, G. W. and Stratton, G. E. (1982). Acoustic communication andreproductive isolation in spiders. In Spider Communication: Mechanismsand Ecological Significance(ed. P. N. Witt and J. S. Rovner), pp. 123-129.Princeton: Princeton University Press.
Vollrath, F. (1979). Vibrations – their signal function for a spiderkleptoparasite. Science205, 1149-1151.
Zahavi, A. (1975). Mate selection: a selection for a handicap. J. Theor. Biol.53, 205-214.
