Conspecific Display Recognition Differs Between Two Species of. Galápagos Lava Lizards: Evidence from Responses to Robots

Transcription

1 *Title Document Conspecific Display Recognition Differs Between Two Species of Galápagos Lava Lizards: Evidence from Responses to Robots David L. Clark a, Joseph M. Macedonia b, John W. Rowe a, Mark A. Stuart a a Department of Biology, Alma College, Alma, MI USA b Department of Biology, Florida Southern College, Lakeland, FL USA Corresponding author: Dr. David L. Clark Alma College - Department of Biology 614 W. Superior St. Alma, MI clarkd@alma.edu Phone: (989) Fax: ( Word count: 4,759

2 Abstract Click here to view linked References Male lava lizards of the Galápagos Islands (Microlophus spp.) exhibit considerable pattern diversity in push-up advertisement displays. No two species overlap in distribution, and all are thought to have evolved in isolation from congeners. Given that push-up display variation in this species-group likely arose as a consequence of genetic drift in allopatry, discrimination of conspecific from heterospecific displays is anticipated to be relaxed. We used programmable lizard robots to test whether two of the species, Microlophus grayii and Microlophus indefatigabilis, could discriminate their own displays from those of a congener. We analyzed the responses of 94 adult male subjects who had been presented with robots possessing conspecific coloration and that performed conspecific or heterospecific push-up displays. Microlophus grayii, a cryptically colored species on Floreana, revealed no evidence of discriminating conspecific from heterospecifc displays. In contrast, M. indefatigabilis, a conspicuously colored species on Santa Cruz, exhibited significantly stronger responses to the conspecific push-up display than to that of the heterospecific. We propose several hypotheses that could explain the differences between these species in social signal recognition, including differential predation pressure, sexual selection, and the possibility of secondary contact between M. indefatigabilis and a neighboring species during periods of glacial maxima. Keywords: Communication, Galápagos Islands, Lizard robots, Microlophus, Push-up display, Sexual selection, Speciation

3 *Manuscript Click here to view linked References Animal signals and displays exhibit extensive variation in structure, modality, and function (Bradbury & Vehrencamp 2011). Signals that facilitate discrimination of conspecifics from sympatric congeners are particularly widespread in vertebrates (Ord & Stamps 2009; Ptacek 2000; Ryan & Rand 1993), and a reduction in fitness resulting from interspecific mating can drive signal differentiation (i.e., reproductive character displacement RCD; Brown & Wilson 1956, Hoskin et al. 2005). For example, when genetically diverging populations come into secondary contact following isolation, character displacement in species recognition signals may be observed along the geographic interface between populations (e.g., Crampton et al. 2011; Lambert et al. 2013; Lemmon 2009; Ng & Glor 2011). Species-specificity in recognition traits can serve as an effective prezygotic reproductive barrier, which is often particularly important for females (Butlin 1987; Servedio 2004). Similarly, displays that facilitate competitor recognition can arise from interspecific interference competition, where agonistic character displacement (ACD) of male phenotypic traits reduces fitness costs associated with inappropriate aggression toward heterospecifics (Grether et al. 2009; Okamoto & Grether 2013). However, in the absence of advantages associated with discrimination of conspecifics from related congeners, as may occur in allopatric speciation, relaxation of interspecific signal discrimination is anticipated (Wellenreuther et al. 2009) Many diurnal lizards perform push-up, or head bob displays that frequently exhibit species-specific patterns (Jenssen 1977). Male lizard displays

4 appear to facilitate species recognition in the contexts of female mate choice and male agonistic competition (e.g., Carpenter & Ferguson 1977; Jenssen 1977; Ord & Stamps 2009; Ord et al. 2011). Yet, lizard species that have evolved in isolation from congeners, such as those in the Galápagos Islands, might be less able to distinguish conspecifics from congeners based on display structure alone The volcanic Galápagos archipelago offers a unique opportunity to study evolution in isolation. Many species found in the Galápagos are endemic and exhibit localized adaptive variation (Parent et al. 2008). Galápagos lava lizards, Microlophus spp., comprise nine species (Benavidas et al. 2009) that vary in body coloration and in push-up advertisement displays (Carpenter 1966; Stebbins et al. 1967; Figure 1). Congeners do not coexist on any island: all nine species are, and may have always been, allopatric (e.g., Benavidas et al. 2009). Although push-up displays are thought to mediate species recognition in many lizard taxa, the allopatric origin of all Galápagos lava lizard species led Carpenter (1966) to suggest that interspecific variation in their displays likely arose by genetic drift. The probability that drift has played a primary role in the phenotypic diversification of Galápagos lava lizards is well supported (e.g., Jordan & Snell 2008). Thus, as each Galápagos Microlophus species is thought to have evolved in isolation from its congeners, selection to maintain discrimination of conspecific from heterospecific display structure would seem unlikely. 45

5 Microlophus indefatigabilis is endemic to the island of Santa Cruz and several nearby islets (Carpenter 1966; Benavidas et al. 2009; Jordon & Snell 2008), where males exhibit striking ventral and lateral orange, yellow, and black advertisement coloration. Although females of all Galápagos Microlophus species display conspicuous orange-red sexual coloration on the head, M. indefatigabilis is unique among these species in that males possess highly chromatic, sexual coloration. In contrast, males of the species Microlophus grayii, which is endemic to the island of Floreana (Benavidas et al. 2009; Carpenter 1966), exhibit a cryptically patterned body of gray, white, brown, and black Experiments using video playbacks or robots have shown that discrimination between conspecific and heterospecific displays occurs in lizards that have evolved in sympatry with congeners (Macedonia & Stamps 1994; Macedonia et al. 2013; Ord & Stamps 2009; Partan et al. 2011). Galápagos lava lizards provide an opportunity to test whether similar discrimination abilities are present in species that presumably have evolved in isolation from congeners In our study, we used lizard robots to test whether Galápagos lava lizards discriminate conspecific from heterospecific push-up displays. For each focal male, we measured the latency to first push-up display, the total duration of push-up displays and a ranked score for aggression based on behaviours described by Carpenter (1966). From the standpoint that each species in this island radiation of Microlophus evolved allopatrically (Benavidas et al. 2009;

6 Carpenter 1966), we hypothesized that species recognition of push-up displays would be weak or absent in these lizards. To test this hypothesis we chose a species pair that differed strongly in their expression of another type of visual signal color ornamentation. We presented subjects with robots that exhibited their own species body color pattern, but which performed either the conspecific or heterospecific push-up display (Experiment 1), or both display patterns sequentially (Experiment 2). The use of robots permitted us to control all attributes of stimulus appearance and behaviour, and allowed us to systematically manipulate the variable of interest: push-up display structure MATERIALS AND METHODS Study Areas We conducted robot presentations to 40 adult male M. grayii between February 25 and February 28, 2012 on Floreana Island. Trials were conducted in the vicinity of Puerto Velasco Ibarra (1º16' 27"S 90º29'13"W) along rock walls and along the roads bordering the village. On Santa Cruz Island, in the vicinity of the Charles Darwin Research Station (0º44'32"S 90º 18'13"W), we presented robots to 54 adult male M. indefatigabilis from February 29 to March 7, Lizard subjects were located on natural lava rock formations and on humanmade structures such as lava rock walls and piles Robot Construction

7 To control robot push-up displays, we secured a Futaba S9001 servomotor (F.I.C. American Corp., Carol Stream IL USA) inside of a plastic housing (25 L x 15 W x 10 H cm) that was painted a dark gray color similar to that of local lava rocks. The servomotor housing also contained a Yost Engineering ServoCenter Midi v1.2 (Portsmouth, OH USA) control board that communicated midi controller messages to the servomotor. A Li-ion battery (Powerizer model # H4HCT , BatterySpace.com/AA Portable Power Corp., Richmond, CA USA) was used to power the robot and control system. In the field, midi controller messages were sent to the input of the ServoCenter from an ipod touch (model #A1213, 1 Infinite Loop, Cupertino CA USA) using a Line 6 MIDI Mobilizer (Line 6, Inc., Calabasas, CA USA) The robot body was constructed from a wooden dowel that we carved to approximate an averaged-sized male lava lizard (74 mm SVL, unpub. data). The body was secured posteriorly via a Micro E/Z hinge (model #DUB937, Du-bro Products, Inc., P.O. Box 815, Wauconda, IL USA) and anteriorly by a small metal eyelet. We attached the eyelet to a pushrod that was connected to the servomotor. When activated, the servomotor controlled the up-and-down push-up display of the robot To produce life-like lizard models, skins for the body were constructed by capturing high-resolution photographs of live lava lizards standing in profile. We

8 removed backgrounds from the digital images using Adobe Photoshop and sized the skin to fit the lizard model. Once a fit for one side of the body was obtained, we created a mirror image to fit the other side of the model. The entire skin was then printed onto a photo-quality, stretchable, sticky-back fabric (Dritz Printed Treasures, Prym consumer USA Inc., Spartanburg, SC) using an inkjet HP Deskjet 460 printer. We cut out the lizard skin image and adhered it to the model. Using a preserved mainland Ecuadorian lava lizard species, Microlophus occipitalis, latex hind limbs and a tail were produced with Plaster-of-Paris impression molds. The latex limbs and tail were painted to approximate the coloration of a live lava lizard and then were glued in place at the posterior portion of the model (online supplement) Push-up Display Pattern Programming Microlophus grayii and M. indefatigabilis each produce species-specific push-up displays, as originally described by Carpenter (1966). Push-up displays are plotted as amplitude X time display action pattern (DAP) graphs (Carpenter & Grubitz, 1961), in which push-ups and pauses are divided into units. During previous work in 2011, we videotaped and transcribed the push-up displays of five adult males of our two study species. A mean DAP graph (Figure 1), generated for each species, was re-constructed as a midi controller file using Logic Pro (v. 9.1 for Macintosh OS) software. A display sequence consisted of three consecutive push-up displays followed by a 30 sec pause, which was iterated for a total trial period of 8 min. To present a display sequence to

9 subjects in field trials, we exported the Logic Pro midi controller file directly to an ipod touch using the Line 6 MIDI Mobilizer Experimental Protocol We searched for adult male lizards ( 65 mm snout-vent length) by walking along paths and dirt roads where lava rocks were present. When a subject was located, the robot was moved to a distance ~1 3 m perpendicular to the lizard s side-on view, so that the robot push-up displays were readily visible to the male. To record the behaviour of the test subject, we secured a digital video camera to a tripod and positioned it ~2 m behind the stimulus robot and in-line with the subject, so that we could capture the robot and the subject in a single field-of-view. If the subject did not flee or display throughout the setup period (ca. 2 min), we triggered the robot s display sequence from an ipod touch, thus initiating the 8 min trial. We avoided searching for lizards in the same areas to avoid testing the same subject more than once (i.e., pseudoreplication) For all 40 M. grayii trials, and for the first 24 of 54 M. indefatigabilis trials, we used a single-presentation technique, where test subjects were presented with a conspecific robot (body color pattern) displaying either conspecific or heterospecific push-ups. We alternated the order of the robot stimuli to control for order effects; thus, each half of our subjects saw one stimulus or the other, but not both. 159

10 During the M. indefatigabilis trials it became apparent that subjects were attending preferentially to the conspecific robot, a response that we had not observed in trials with M. grayii males. The seemingly greater attention of M. indefatigabilis males for the conspecific display inspired us to implement a matched-pairs ( stimulus switch ) design for 30 male subjects. Here, each subject was presented with a conspecific as well as a heterospecific display. We alternated sequentially which display type (conspecific or heterospecific) a given subject witnessed first in a trial. Thus, 15 subjects were presented with 8 min of conspecific display followed by 8 min of heterospecific display, and 15 subjects observed 8 min of heterospecific display followed by 8 min of conspecific display Description of Behaviours Responses to robot stimuli by male lava lizard subjects were quantified directly from video recordings. The Push-up Display is a visual signal used by males in the context of territory advertisement and defense, and to attract female conspecifics (e.g., Carpenter & Ferguson, 1977). In each robot presentation trial we scored male subjects for total Push-up Display duration (sec) as well as latency to begin these displays We also developed a ranked composite score for subjects responses that was based on lava lizard aggressive behaviours originally described by Carpenter (1966). Our Composite Response score (rank value indicated in parentheses below) consisted of behaviours that may be exhibited together or

11 independently: Crest Up (1 point) the nuchal and dorsal crests are raised exposing brightly colored scales; Gular Inflation (1 point) the neck region becomes inflated and distended; Lateral Compression (1 point) the male turns lateral to the stimulus and becomes rigid and flattened; and Challenge Display (4 points) all three of the behaviours described with the addition of an archedback Push-up Display (see online Video 1). When computing a subject s scores, the maximum score allowed was 4; i.e., if a subject escalated to the Challenge Display, points from engaging in the other three behaviours described above were not added to the total score. Durations of Push-up Displays exhibited during the Challenge Display were included in the Push-up Display total for a trial. Male lava response to robot stimuli and typical field set up are shown in online Videos 2 and Statistical Analysis The durations of trials varied among individuals due to some subjects moving out of view or fleeing. We therefore converted a subject s total duration of Push-up Display to a proportion of the stimulus presentation period during which the subject was clearly visible in the video camera footage, or before the subject fled and did not display again. As transformations failed to normalize Push-up Display measurements, we tested for differences in responses to conspecific and heterospecific display stimuli using nonparametric statistical tests. Wilcoxon 2-Sample Tests were conducted on the mean Push-up Display latency (secs), the total Push-up Display duration proportion of the trial, and the arcsine

12 transformed Composite Response scores using JMP v9.0. For the stimulusswitch trials, responses of subjects were tested using the Wilcoxon Matched- Pairs Signed-Rank test. Responses were tested first for effects of stimulus order (irrespective of stimulus display pattern), and then were tested for effects of stimulus display pattern (irrespective of stimulus order). As the same data set from the stimulus-switch trials was used to test for effects of stimulus type and stimulus order, we adjusted our alpha level for significance to α = RESULTS Single Stimulus Presentations Microlophus grayii. Male M. grayii exhibited statistically indistinguishable responses to conspecific and heterospecific display stimuli. First, we failed to detect a difference in mean latency to the first push-up display by M. grayii subjects in response to conspecific and heterospecific displays (Wilcoxon 2-Sample Test: X21 = , P > 0.94; Fig. 2a). Likewise, we found no significant difference in the mean proportion of push-up displays per trial in response to the two different species displays (M. grayii display: X +SE = ; M. indefatigabilis display: X +SE = ; Wilcoxon 2-Sample Test: X21 = , P > 0.56; Fig. 3a). Last, we did not find a significant difference in mean Composite Response score between conspecific and heterospecific displays (M. grayii display: X +SE = ; M. indefatigabilis display: X +SE = ; Wilcoxon 2-Sample Test: X21= 0.058, P > 0.80). 228

13 Microlophus indefatigabilis. In contrast to M. grayii males, M. indefatigabilis males exhibited significantly stronger responses to conspecific than to heterospecific push-up displays. First, mean latency to the first push-up display was significantly shorter in response to conspecific than to heterospecific displays (Wilcoxon 2-Sample Test: X21 = 3.97, P < 0.04; Fig. 2b). Similarly, the mean proportion of time spent in push-up displays during a trial was significantly greater in response to conspecific than to heterospecific displays (M. indefatigabilis display: X +SE = ; N = 12; M. grayii display: X +SE = ; N = 12; Wilcoxon 2-Sample Test: X21= 5.07, P < 0.02; Fig. 3b). Finally, the mean Composite Response score of lizards was significantly greater for males presented with the conspecific display compared to males presented with the heterospecific display (M. indefatigabilis display: X +SE = , N = 12; M. grayii display: X +SE = , N = 12; Wilcoxon 2-Sample Test: X21 = 3.99, P < 0.046; N = 12) Stimulus-Switch Presentations M. indefatigabilis Although subjects responded more strongly to conspecific than to heterospecific displays, we were unable to detect an effect of stimulus order on push-up duration response (Wilcoxon Signed Rank Test, S = 24.50, N = 30, P > 0.62; Fig. 4a,b). However, data analysis revealed an effect of stimulus type, where mean total push-up duration per trial was significantly greater in response to the conspecific than to the heterospecific displays (Wilcoxon Signed Rank Test, S = , N = 30, P < ; Fig. 4a,b).

14 DISCUSSION Elaborate color ornamentation and motion displays of animals have long been considered in light of sexual selection theory (i.e. Andersson 1994; Darwin 1871). Selection seems likely to favor the production and recognition of speciesspecific displays when the potential exists for interspecific mating that leads to reduction in fitness (i.e. reproductive character displacement: Butlin 1987; Servedio 2004), or when species recognition signals reduce inappropriate aggression towards heterospecifics (i.e. agonistic character displacement: Grether et al. 2009). Recent use of robotic lizard stimuli has demonstrated that dewlap color and headbob display patterns contribute independently to species recognition in Anolis grahami a species that has evolved within a community of sympatric, closely related species (Macedonia et al. 2013). Given that Galápagos lava lizards are argued to have evolved in allopatry (e.g., Benavidas et al. 2009; Carpenter 1966), we hypothesized that maintenance of species recognition should be relaxed in these lizards, i.e., discrimination should be weak or absent. Interestingly, although our results are consistent with this hypothesis in one of our study species (M. grayii), they run counter to it in our other study species (M. indefatigabilis) For M. grayii, a cryptically color-patterned species, mean latency to the first push-up, push-up duration per trial, and Composite Response scores were virtually identical for subjects responses to conspecific and heterospecific

15 displays. This outcome suggests that species recognition of ritualized displays may indeed be relaxed for animals that evolve in isolation, as has been found elsewhere (e.g., Wellenreuther et al. 2009). In contrast, M. indefatigabilis, a more ornate and colorful species, exhibited clear discrimination between conspecific and heterospecific displays for the responses that we measured. Given that the responses of M. indefatigabilis males to conspecific and heterospecific displays were so strikingly different from those of M. grayii, we designed an additional experiment to confirm the results of our first experiment. In this second experiment, push-up display pattern was switched mid-trial from one species display to the other. Whereas we did not detect an effect of presentation order, responses to the conspecific display differed dramatically from those elicited by the heterospecific display. The effect of changing the display mid-trial was often quite conspicuous to us for subjects that were shown the heterospecific display first. In such cases, within a few seconds of the switch to the conspecific display, subjects that had not been responding or who were responding weakly often would fixate visually on the robot, exhibit an alert and aggressive posture, move toward the robot, and begin to engage in push-up displays The disparate experimental results for M. grayii and M. indefatigabilis, both of which presumably have evolved in allopatry, lead us to hypothesize how selective pressures might have differed in these two species to result in such dramatic differences in push-up display pattern recognition. Although the

16 possibility that substantially greater predation pressure on Floreana than on Santa Cruz is consistent with differences in male body coloration (Fig. 1 and online supplements), we are unaware of any published work on predation pressure experienced by M. grayii and M. indefatigabilis. Such research could be enlightening, however. For example, in southern New Mexico where predator diversity and abundance is high and several key predators are present, the common collared lizard (Crotaphytus collaris) possesses body coloration that is exceedingly cryptic; by contrast, in eastern Utah where predator diversity and abundance is low and the same key predators are absent, body coloration in the same species is highly conspicuous (Macedonia et al. 2002). Perhaps a parallel situation exists for our Microlophus species in the Galápagos. Whether or not predation pressure can explain body color differences between male M. grayii and male M. indefatigabilis, males of the latter species are by far the most color-ornamented males of any Microlophus species in the Galápagos. We speculate that push-up pattern recognition might arise as a correlated byproduct of heightened discrimination associated with condition dependent coloration (e.g., an indicator trait: Grether 2010). We are uncertain at present how this indicator trait hypothesis might best be tested, but one prediction arising from it would be the absence of conspecific-heterospecific display discrimination in males of unornamented Galápagos Microlophus spp. a prediction already supported by our data for M. grayii. An alternative reproductive reinforcement hypothesis proposes that display recognition in M. indefatigabilis stems from the evolution of pre-mating

17 reproductive isolation during secondary contact with Microlophus albemarlensis on Isabella Island during Pleistocene glacial maxima. Until relatively recently, M. indefatigabilis was not considered a distinct species from M. albemarlensis (e.g. Kizirian et al. 2004). Moreover, it has been suggested that Santa Cruz and Isabella islands may have been connected by a land bridge during periods of unusually low sea levels (Geist et al. in press). One testable prediction of this hypothesis is that, when subjected to a robot push-up display choice test, males of the cryptically colored M. albemarlensis will respond preferentially to their own species push-up displays compared to those of M. indefatigabilis. Future work will be aimed at utilizing our robots to test this hypothesis with M. albemarlensis, as well as to explore female preferences for social and sexual signals in the Galápagos radiation of lava lizards

18 REFERENCES Andersson, M. (1994). Sexual selection. Princeton, NJ: Princeton University Press Benavides, E., Baum, R., Snell, H. M., Snell, H. L., & Sites, J. W. (2009). Island Biogeography of the Galapagos Lava Lizards (Tropiduridae: Microlophus): Species Diversity and Colonization of the Archipelago. Evolution, 63-6: (DOI: /j x) Bradbury, J. W., & Vehrencamp, S. L. (2011). Principles of animal communication. Sunderland, MA: Sinauer Associates, Inc Brown, W. L., & Wilson, E. O. (1956). Character displacement. Systematic Zoology, 5, Butlin, R. (1987). Speciation by reinforcement. Trends in Ecology & Evolution, 2, Carpenter, C. C., & Grubitz, G. G. (1961). Time-motion study of a lizard. Ecology, 42, Carpenter, C. C. (1966). Comparative behaviour of the Galápagos lava lizards (Tropidurus), p (Paper 35). In R. I. Bowman (Ed.) The Galápagos:

19 Proceedings of the Galápagos international scientific project. Univ. California Press, Berkeley Carpenter, C. C., & Ferguson, G. W. (1977). Variation and evolution of stereotyped behaviour in reptiles. In C. Gans (Ed.), Biology of the reptilia, 18 (pp ). Chicago: University of Chicago Press Crampton, W. G. R., Lovejoy, N. R., & Waddell, J. C. (2011). Reproductive character displacement and signal ontogeny in a sympatric assemblage of electric fish. Evolution, 65, DOI: /j x Darwin, C. (1871). The descent of man, and selection in relation to sex. London, U.K., John Murray Geist, D., Snell, H., Snell H., Goddard, C., & Kurz, M. (in press). Paleogeography of the Galápagos islands and biolgeographical implications. AGU Geophysical Monograph Grether, G. F., Losin, L., Anderson, C. N., & Kenichi, O. (2009). The role of interspecific interference competition in character displacement and the evolution of competitor recognition. Biological Reviews, 84,

20 Grether, G. F. (2010). The evolution of mate preferences, sensory biases, and indicator traits. In: H. J. Brockmann (Ed.), Advances in the Study of Behaviour, 41, (pp ). Burlington: Academic Press Hoskin, C. J., Higgie, M., McDonald, K. R., & Moritz, C. (2005). Reinforcement drives rapid allopatric speciation. Nature, 437, DOI: /nature Jenssen, T. A. (1977). Evolution of anoline lizard display behaviour. American Zoologist, 17, Jordan, M. A., & Snell, H. L. (2008). Historical fragmentation of islands and genetic drift in populations of Galápagos lava lizards (Microlophus albemarlensis complex). Molecular Ecology, 17, DOI: /j X x Kizirian, D., Trager, A., Donnelly, M. A., & Wright, J. W. (2004). Evolution of Galápagos lava lizards (Iguania: Tropiduridae: Microlophus). Molecular Phylogenetics & Evolution, 32, Lambert, S. M., Geneva, A. J., Mahler, D. L., & Glor, R. E. (2013). Using genomic data to revisit an early example of reproductive character displacement in Haitian Anolis lizards. Molecular Ecology, 22, (DOI: /mec.12292)

21 Lemmon, E. M. (2009). Diversification of conspecific signals in sympatry: geographic overlap drives multidimensional reproductive character displacement in frogs. Evolution, 63, (DOI: /j x) Macedonia, J. M., & Stamps, J. A. (1994). Species recognition in Anolis grahami (Sauria, Iguanidae): Evidence from responses to video playbacks of conspecific and heterospecific displays. Ethology, 98, Macedonia, J.M., Brandt, Y.M., & Clark, D.L. (2002). Sexual dichromatism, adaptive colouration and differential conspicuousness in two populations of the collared lizard, Crotaphytus collaris. Biological Journal of the Linnean Society, 77: Macedonia, J.M., Clark, D.L., Riley, R., & Kemp, D.J. (2013). Species recognition of color and motion signals: Evidence from responses to lizard robots. Behavioural Ecology, 24: (DOI: /beheco/art027) Ng, J., & Glor, R. E. (2011). Genetic differentiation among populations of a Hispaniolan trunk anole that exhibit geographical variation in dewlap color. Molecular Ecology, 20, (DOI: /j X x) 432

22 Okamoto, K. W., & Grether, G. F. (2013). The evolution of species recognition in competitive and mating contexts: the relative efficacy of alternative mechanisms of character displacement. Ecology Letters, 16, (DOI: /ele.12100) Ord, T. J., & Stamps, J. A. (2009). Species identity cues in animal communication. Amercan Naturalist, 174, (DOI: /605372) Ord, T. J., King, L., & Young, A. R. (2011). Contrasting theory with the empirical data of species recognition. Evolution, 65, (DOI: /j x) Parent, C. E., Caccone, A., & Petren, K. (2008). Colonization and diversification of Galápagos terrestrial fauna: a phylogenetic and biogeographical synthesis. Philosophical Transactions of the Royal Society B, 363, (DOI: /rstb ) Partan, S. R., Otovic, P., Price, V. L. & Brown, S. E. (2011). Assessing display variability in wild brown anoles Anolis sagrei using a mechanical lizard model. Current Zoology, 57, Ptacek, M. B. (2000). The role of mating preferences in shaping interspecific divergence in mating signals in vertebrates. Behavioural Processes, 51:

23 Ryan, M. J., & Rand, A. S. (1993). Species recognition and sexual selection as a unitary problem in animal communication. Evolution, 47, Servedio, M. R. (2004). The evolution of premating isolation: local adaptation and natural and sexual selection against hybrids. Evolution, 58: (DOI: /j tb00425.x) Stebbins, R. C., Lowenstein, J. M., & Cohen, N. W. (1967). A Field Study of the Lava Lizard (Tropidurus albemarlensis) in the Galápagos Islands. Ecology, 48, Wellenreuther, M., Tynkkynen, K., & Svennson, E. I. (2009). Simulating range expansion: male species recognition and loss of premating isolation in damselflies. Evolution, 64, (DOI: /j x) 470

24 FIGURE LEGENDS Figure 1. Comparison of morphology and Display Action Pattern of Galápagos Islands lava lizards a) M. grayii; b) M. indefatigabilis Figure 2. Comparison of mean latency to first push-up display by a) M. grayii and b) M. indefatigabilis in response to a conspecific and a heterospecific robot push-up display pattern Figure 3. Comparison of mean push-up display duration (seconds) by a) M. grayii; b) M. indefatigabilis when presented with a conspecific and a heterospecific robot push-up display pattern Figure 4. Comparison of mean push-up display duration (seconds) by M. indefatigabilis to a robotic lizard that displayed a) the conspecific display first; and b) the heterospecific display first.

25 *Highlights (for review) Highlights Species that evolve in allopatry might show relaxed species recognition capabilities We tested species recognition between two species of Galápagos lava lizards One species did not discriminate while the other species was highly discriminating Sexual selection or secondary contact could explain differences among species

26 Acknowledgments ACKNOWLEDGEMENTS We thank W. Tapia and G. Quezada of the Galápagos National Park for permission to conduct this research on Santa Cruz and Floreana Islands. We also wish to thank Sonia Cisneros, Roberto Robby Pepolas Lecaro and the staff at the Charles Darwin Research Station for their help in facilitating our research. We are grateful for support that Alma College obtained from the MacGregor Foundation, and to R. Clark and K. Uetz for technical advice concerning robot construction.

27 Figure 1. Click here to download high resolution image

28 Figure 2. Mean Latency to First Push- up Display (secs) a. b. M. grayii M. indefatigabilis Species Tested Robot Display M. grayii M. indefatigabilis Figure 2.

29 Figure 3. Mean Push-up Display Duration (secs) a. b. M. grayii M. indefatigabilis Species Tested Robot Display M. grayii M. indefatigabilis Figure 3.

30 Figure 4. Mean Push-up Display Duration (secs) a. b Robot Display M. grayii M. indefatigabilis 10 0 First Second First Second Order of Stimulus Presentation Figure 4.

31 Video Still - Challenge Display Click here to download high resolution image

32 Video Still - Lava lizard response to robot (M. grayii) Click here to download high resolution image

33 Video Still - Lava lizard response to robot (M. indefatigabilis) Click here to download high resolution image