Juvenile Collared Lizards Adjust Tail Display Frequency in Response to Variable Predatory Threat

Ethology Juvenile Collared Lizards Adjust Tail Display Frequency in Response to Variable Predatory Threat Joshua R. York* & Troy A. Baird * Department of Biology, University of Oklahoma, Norman, OK, USA Department of Biology, University of Central Oklahoma, Edmond, OK, USA Correspondence Joshua R. York, Department of Biology, University of Oklahoma, 730 Van Vleet Oval, Norman, OK 73019, USA. E-mail: joshuayork@ou.edu Received: May 15, 2015 Initial acceptance: June 6, 2015 Final acceptance: October 12, 2015 (L. Ebensperger) doi: 10.1111/eth.12442 Keywords: antipredator behavior, antipredator display, lizard tail displays, predator avoidance behavior, predator deflection, predator deterrence Abstract Antipredator behavioral tactics have evolved in phylogenetically diverse animal clades and often involve prey initiating conspicuous display patterns when encountering potential predators. Using detailed behavioral observations in the field, we report the first description of conspicuous tail displays in juvenile collared lizards (Crotaphytus collaris), a species that does not have tail autotomy. When approached in the field, lizards gave four stereotypical displays involving the tail, suggesting that these might function as antipredator signals. To test this hypothesis, we compared the frequency of tail displays given by juveniles during approach experiments relative to control trials involving observation from afar. We then further manipulated the intensity (low threat, high threat) with which we threatened lizards by varying our angle of approach relative to the body-axis orientation of lizards resting on a perch sloping away from the observer. Our results show that juveniles consistently performed all four types of tail displays in response to our approaches, but subjects never displayed during non-threat control trials. Moreover, lizards were more likely to remain emergent and give tail displays when we approached them head-on with their bodies sloping away (low threat), but were more likely to take refuge when approached from the side and behind (high threat). The positive relationship between the frequency of display and risk level suggests that tail displays function to signal that juvenile lizards have detected potential predators, which may deter further pursuit. Together, our results provide the first account of visual displays involving the tail in collared lizards and suggest that these displays function to signal potential predators. Introduction A wide variety of behavioral traits are shaped by natural selection for predator avoidance, including those involved in escaping or minimizing attack (Greene 1988; Caro 2005; Stankowich & Blumstein 2005). Prey sometimes simply flee and/or take refuge when they encounter potential predators (Broom & Ruxton 2005). Although fleeing and hiding may decrease the likelihood of predation, these responses also may compromise the ability of prey to engage in other activities that promote fitness, such as remaining motionless for sit-and-wait foraging (Martın & Lopez 1999; Amo et al. 2007), as well as opportunities to mate or thermoregulate. The apparent selective tradeoff between predator avoidance and activities that require exposure to attack may promote the evolution of antipredator behavioral tactics that balance vulnerability against other essential demands of fitness (Ydenberg & Dill 1986; Ruxton et al. 2004). Such tactics have evolved independently in phylogenetically diverse animal clades, and often involve prey directing displays to approaching predators even though they appear to render prey more conspicuous and apparently more vulnerable (Cooper et al. 2004; Langkilde et al. 2004). For example, several species of Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH 37

Antipredator Tail Displays in a Lizard J. R. York & T. A. Baird fish, mammals, and insects are known to advertise colorful/ornamented body parts to potential predators, which may deter further predatory pursuit (Caro 2005; Stankowich & Blumstein 2005). Similarly, several birds and amphibians produce acoustic signals in response to predatory threat to alert predators that they have been detected (Caro 2005; Stankowich & Blumstein 2005). Lizards have emerged as excellent models for studying the adaptive significance of antipredator behavioral signals. Many lizard species give highly conspicuous displays, most often involving the tail, when threatened by predators in both natural habitats and laboratory settings (Greene 1988; Hasson et al. 1989; Telemeco et al. 2011). Three primary mechanisms have been hypothesized to explain the evolution of conspicuous, apparently risk-prone tail displays that are given by lizards. The simplest explanation is that tail displays evolved to deflect attacks to a non-essential body part (Bateman & Fleming 2009; Cooper & Vitt 2010). Tail displays are particularly likely to evolve through deflection in lizards having highly ornamented tails that autotomize when grasped affording prey sufficient time to escape while the predator consumes the tail (Cooper 1998; Telemeco et al. 2011). Alternatively, especially in lizards that lack tail autotomy, tail displays may communicate more detailed content that functions to deter pursuit by predators (Hasson et al. 1989; Cooper et al. 2004). For example, prey may signal that they have detected approaching predators (perception advertisement hypothesis, sensu Caro 1995), causing predators to abandon further pursuit. Tail displays may also signal that prey are capable of evading an impending attack because nearby refuges are abundant, or from their current position relative to the predator, prey are physically capable of escape by fleeing (relative escape ability hypothesis, sensu Caro 1995). During routine mark recapture experiments, we observed that juvenile collared lizards (Crotaphytus collaris) sometimes moved their tails in a stereotypical manner when we approached them, suggesting that these displays may function as antipredator signals. Because these displays have not been previously reported in this species, we first give a detailed description of the display patterns involving the tail. We then experimentally tested the hypothesis that these stereotypical tail displays function as antipredator signals by (1) approaching free-ranging juveniles to simulate a predatory threat (Cooper 2003) and recording tail display activity and (2) manipulating the intensity with which we threatened lizards by varying the angle of approach. Materials and Methods Study Site and Population Collared lizards are diurnal medium-sized Iguanians that occur throughout the central and southwestern USA. Natural habitat for collared lizards consists of exposed rock outcroppings and washes, but they have also colonized human constructed habitats composed of hard substrates such as flood control spillways associated with dams (Baird 2013a,b; Husak 2006a, b). Collared lizards use elevated rock perches to bask and scan for arthropod prey, and they take refuge from potential predators and excessive heat in crevices beneath or adjacent to these perches (Baird & Sloan 2003; Baird et al. 2003; Baird 2013a). This study was conducted at the Arcadia Lake Dam located 9.6 km east of Edmond, OK, USA, which is well suited for observation of collared lizard behavior because human access is restricted and the lizards are undisturbed (Baird 2013a). We conducted experiments on two habitat patches (380 9 3 m and 740 9 3 m) composed of contiguous pieces of broken concrete slab. These patches are optimal for observation of lizard responses to manipulations of simulated predation threat because they are narrow, unobstructed, and bordered on both sides by grass. Because the broken pieces of concrete (0.25 0.75 m) rest at variable angles, they provided abundant slanted perches and allowed us to approach lizards from carefully controlled angles which varied the degree to which lizards were exposed to the observer. Grounddwelling predators of collared lizards at our study site are coachwhips (Masticophis flagellum), roadrunners (Geococcyx californianus), and coyotes (Canis latrans), whereas raptors (Ictinia mississippiensis, Buteo jamaicensis) attack lizards from above. Experimental Subjects and Protocol From mid-july through late October 2011 and 2012, we surveyed the entire study site daily to capture (by noose), mark, and measure newly emerged hatchlings, and then recapture when they molted. At each capture, we recorded snout-to-vent length (1 mm) and total body mass (0.1 g). Lizards were marked permanently using unique toe-clips and identified from a distance by applying non-toxic paint spots to the dorsum. Newly emerged hatchlings in mid-july and early August were 38.0 40.0 mm SVL and had approximately doubled in size (58.0 99.0 mm SVL) by the end of October. For our experiments, we used 38 Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH

J. R. York & T. A. Baird Antipredator Tail Displays in a Lizard juveniles that were approximately 56.0 100 mm SVL. We only used juveniles that had full tails, avoiding those having part of their tails bitten off by predators (collared lizards lack tail autotomy septa). All experiments were conducted from 0900 to 1300 h when the substrate temperature was 30 38 C. Over the 8 C temperature range that we ran experiments, collared lizard activity, including escape behavior, is not dependent on surface temperature (Baird et al. 2001; Braun et al. 2010). To control for possible effects of lizard habituation to observers, all subjects were previously captured, marked, and measured no more than twice prior to experiments. For all trials, we first identified subjects perched on elevated rocks from a distance 30 40 m using their unique color codes. The observer (JRY) approached subjects holding a noose pole, using one of three techniques to vary threat level categorically (non-threat, low threat, high threat; described below). For all treatments, we recorded the number of seconds that lizards spent giving tail displays for 10 min, and calculated the percent time per trial that lizards displayed using their tails. To test the hypothesis that tail displays function as antipredator signals, we first compared all tail activity by subjects (n = 12) during non-threat versus lowthreat conditions. The non-threat treatment involved the observer sighting lizards from afar and, once the observer had approached within 10 m of the subject, stopped and observed subjects without further approach (Fig. 1). For the low-threat treatment, we chose subjects that were resting on an elevated rock perch that sloped away from the observer, such that the lizard s median axis was oriented parallel with the observer s line of approach (Fig. 1), and only the head (and elevated tail) was exposed to the simulated predator. Subjects were approached using a constant pace (1.5 m/s; Cooper 2003). When subjects first moved, the observer paused and recorded whether lizards immediately took refuge in an adjacent crevice without giving tail displays, or responded by displaying using the tail while remaining emergent. If lizards gave tail displays, we characterized them (see Characterization of tail displays) and recorded the duration of each (1.0 s). When lizards remained emergent in response to approach and ceased displaying, the observer resumed approach until subjects either displayed again, or took refuge in a crevice and did not re-emerge. Each subject lizard was used in both the non- and low-threat treatments in random order. Trials on the same subjects were conducted no sooner than 1 d following their previous trial. Fig. 1: Experimental protocol for non-threat (open arrow), low-threat (hatched arrow), and high-threat (solid arrows) trials. All trials began with the observer 10 m away from a subject perched on an elevated rock. For both non-threat and low-threat trials, the observer was facing the lizard s head, parallel with the subject s midline. High-threat trials began behind and to the side of subjects 160 180 degrees relative to the direction of the subject lizard s head (solid arrows). We further tested our hypothesis that tail displays by juvenile collared lizards function as antipredator signals by pooling the initial 12 low-threat trials with an additional 43 low-threat trials (total low-threat trials, n = 55 using the same protocol) for comparison with trials involving high threat (n = 38). High-threat trials involved the observer approaching perched lizards 160 180 degrees to either side of the direction that the subject s head was facing (Fig. 1). Following Cooper (2008), we reasoned that these angled approaches from behind and laterally posed an increased threat because the entire body of the lizard was exposed to the simulated predator relative to only the head being exposed during the low-threat treatment, and the ability of lizards to visually evaluate the speed and distance of the predator when approached from behind and laterally would be compromised. Although the initial 12 low-threat trials were conducted before high-threat trials, over the entire sample of low- and high-threat trials, the order of treatments was random. Statistical Analyses For all analyses, we used generalized linear mixed models (GLMM; package lme4) in the program R v.3.0.1 (R Development Core Team 2013). We report effect sizes for our categorical (Cohen s d) predictor variables, and considered effects Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH 39

Antipredator Tail Displays in a Lizard J. R. York & T. A. Baird statistically significant when the associated 95% confidence intervals (CI) did not include zero (Colegrave & Ruxton 2003; Nakagawa & Cuthill 2007). We interpreted the relative strength of effect sizes from resulting models according to Cohen (1992) (small effect = below 0.2; moderate effect = 0.2 0.5; large effect =>0.50). To test whether tail displays in juvenile collared lizards function as antipredator signals, we used threat level (non-threat versus low threat) during experiments as the categorical predictor variable in a paireddesign GLMM (Gaussian error structure, identity link function), with lizard identity as the random effect to control for multiple observations on some individuals. We used a GLMM (Gaussian error structure, identity link function) with threat level (low threat versus high threat) as the categorical predictor variable to test whether or not variation in the intensity of threat influenced the cumulative percent time per trial that juveniles gave all types of tail displays, and included lizard identity as a random effect. Finally, we used a GLMM (binomial error structure, log link function) to compare the probability that lizards stayed emergent and gave tail displays versus seeking refuge without displaying in low-threat versus high-threat trials. For this model, we used threat level (high threat, low threat) as the categorical predictor variable and lizard response (remain emergent and display or seek refuge immediately without displaying) as the dependent variable, with lizard identity as a random effect. We limited our analyses to the cumulative percent time displaying for all types of displays pooled, because there was no obvious pattern with which lizards gave the four display types in response to different threat treatment levels. and Use Committee at the University of Central Oklahoma (permit number 13009) and the Oklahoma Department of Wildlife (permit number 5553). We have conducted 25 consecutive seasons of longitudinal studies on collared lizard behavior, growth, and survival in this population. These have involved clipping the terminal phalanges of three digits, application of non-toxic acrylic paint to the dorsum, repetitive capture by noosing, and making body size measurements. Monitoring the behavior and survival of lizards has confirmed that these techniques have had no adverse effects on their health (Baird 2013a; York et al. 2014; York & Baird 2015). Results Characterization of Tail Displays Juvenile collared lizards performed four variations of displays using their tails. The most common display involved conspicuous elevation of the straightened tail above the substrate and moving it back and forth laterally (=lateral wave, display just beginning in Fig. 2a, lateral movement indicated by the arrows). Lizards also frequently displayed by elevating the tail and moving it using a stereotypical slow and deliberate sinusoidal motion (=sinusoidal wave, Fig. 2b). Less frequently, lizards raised and held the tail stiffly in two different positions. Tail curl (shown just beginning in Fig. 2c) involved slowly raising and curling the distal portion of the tail until it was directly above the head and pointed anteriorly. Tail raise involved rapidly elevating the tail and holding it stiffly above the substrate (same as in Fig. 2a, but without lateral movement). Ethical Note All procedures performed on live lizards were conducted with approval of the Institutional Animal Care Tail Display under Different Threat Levels Juvenile collared lizards did not initiate tail displays during non-threat control trials (n = 12), but they (a) (b) (c) Fig. 2: Tail displays given by juvenile collared lizards. (a) lateral wave (lateral motion of the tail indicated by arrows) or tail raise while it is held straight and stiff (without arrows); (b) sinusoidal wave; (c) tail curl just beginning in the photograph, but from here will ultimately be raised over the head. Photography by Teresa D. Baird. 40 Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH

J. R. York & T. A. Baird Antipredator Tail Displays in a Lizard gave displays frequently (all four displays pooled) during 10 of 12 (83%) low-threat trials (t 2, 22 = 5.43, Cohen s d = 2.22, 95% CI = 1.14 2.29, Fig. 3). The average percent time spent giving tail displays during low-threat trials was 3.9 times higher (t 2, 32 = 3.17, Cohen s d, 95% CI = 0.63, 0.23 1.03) than that during high-threat trials (Fig. 4). Moreover, in highthreat trials lizards were over 3.5 times more likely to immediately seek refuge without displaying in one of the many rock crevices adjacent to (<0.5 m) to their perches (t 2, 32 = 4.48, Cohen s d, 95% CI = 0.81, 0.4 1.21) compared with lizards in lowthreat trials that more frequently remained emergent and displayed. Fig. 3: The percentage of time spent tail displaying during non-threat versus low-threat trials. Data are means 1.0 SE. Asterisk indicates a statistically significant effect (95% CI did not included zero). Fig. 4: The percentage of time spent tail displaying during low-threat versus high-threat trials. Data are means 1.0 SE. Asterisk indicates a statistically significant effect (95% CI did not included zero). Discussion Juvenile collared lizards consistently gave tail displays in response to approach by a simulated predator, but never gave these displays during non-threat trials. Moreover, during low-threat trials, subjects increased the frequency of tail display, whereas they usually sought refuge without displaying when approached from angles of increased predatory threat. When subjects were exposed to lower threat because they were more hidden and less able to judge speed and distance of the simulated predator, the costs of signaling to potential predators were likely reduced (Cooper 2008). By contrast, increased exposure during highthreat trials apparently made the cost of displaying prohibitively high, prompting lizards to instead take refuge without displaying. The selective trade-off between threat level and the decision to display is likely acute in species with highly developed vision such as collared lizards (Fox & Baird 1992), which can likely use visual cues of the rate and directionality of approach by predators to assess the relative intensity of threats (Burger & Gochfeld 1981; Braun et al. 2010). Nevertheless, it remains possible that other variables that we could not control during our experiments on free-ranging lizards (e.g., proximity to refuges, conspicuousness/vulnerability to predators) may have influenced perceptions of threat intensity, and hence decisions to display or take refuge. Our results support the hypothesis that tail displays given by juvenile collared lizards function as antipredator signals, similar to displays given by other lizards when threatened (Cooper et al. 2004; Font et al. 2012). Based on our experiments, we cannot unequivocally accept or reject any of the three hypothesized mechanisms proposed to explain how tail displays function in predator avoidance. One possibility is that tail displays deflect attacks to a nonessential body part. In the majority of lizards and other taxa in which displays function for deflection, the non-essential body parts are conspicuously colored and these species possess autotomy septa that suddenly sever the tail when it is struck affording prey the opportunity to escape (Hill & Vaca 2004; Telemeco et al. 2011). Collared lizards lack both conspicuous caudal coloration and autotomy septa. Nevertheless, the tail remains an expendable body part because it is not essential for immediate survival. Although the absence of both conspicuous caudal coloration and autotomy makes it unlikely that the deflection hypothesis explains the evolution of tail displays in juvenile collared lizards, we cannot rule out this possibility entirely. Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH 41

Antipredator Tail Displays in a Lizard J. R. York & T. A. Baird It also seems unlikely that tail displays advertise relative escape ability in our population. Lizard escape ability is often estimated by measuring sprint speed in the laboratory on flat straight tracks. Lizards are elicited to run by prodding them, and maximum sprint speed is measured as that attained at the end of the track (Husak 2006a, b; Husak & Fox 2006; Telemeco et al. 2011). Because we did not measure sprint speed, we cannot rule out a possible relationship between sprinting ability and tail display behavior, but this seems unlikely because it is not necessary for collared lizards at our site to run to access refuges. Because crevices are very abundant throughout our study site (Baird & Sloan 2003; York & Baird 2015), study subjects escaped by darting into refuges that were immediately adjacent (within 0.5 m) on at least two sides of chosen perches, a distance so short that maximal sprint speed is probably not even possible to attain. Other performance traits such as reaction time and the ability to move over very short distances into a crevice may promote escape behavior in collared lizards. However, we are not aware of any published data on such traits in lizards or any other vertebrates. Taken together, our current results suggest that the most likely adaptive explanation of increased tail display frequency in response to low levels of threat is to signal predators that they have been detected (perception advertisement, sensu Caro 1995). Traits that evolved originally for one function are sometimes selectively co-opted for different functions (Bock 1958; Gould & Vrba 1982; Endler & Basolo 1998), and this may have occurred in the case of antipredator displays in some species. For example, tail displays that evolved initially to promote intraspecific communication may subsequently take on an antipredator signaling role (Johnson & Wade 2010). Intraspecific selection associated with sexual behavior is not likely in reproductively immature juvenile collared lizards. However, observations of juvenile collared lizards performing sinusoidal tail movements while scanning the surrounding grass for insect prey and tail curling displays when juveniles were stalking grasshoppers both suggest that tail displays may also play a role in foraging (Braun & Baird unpublished data). Effective sit-and-wait foraging requires that lizards remain exposed on elevated perches for extended periods, whereas departing good foraging sites to take refuge disrupts opportunities to ambush prey (Scarratt & Godin 1992; Perez-Tris et al. 2004). Interruption of foraging activities would be especially costly for juvenile lizards because of their high energetic demands to support rapid growth and to build energy stores to survive the winter (Baird 2008; Braun et al. 2010). Tail displays that involve deliberate tail movements and postures may visually distract prey just long enough to give lizards a momentary advantage when they strike (see similarly, Baird 2008). Because ambush foraging requires sitting exposed on perches such that lizards are vulnerable, it is entirely possible that tail displays that evolved originally to deter predators may have been co-opted later to improve foraging efficiency through distraction of prey, or vice versa. Acknowledgments We thank Bill Parkerson of the US Army Corps of Engineers for permission to conduct field studies at the Arcadia Lake site. The Office of Research and Grants at the University of Central Oklahoma provided funding to JRY and TAB. We also thank Teresa Baird, Sharon LaFave, Will Unsell, and Emily York for providing technical assistance. Permission to conduct this research was provided by the Oklahoma Department of Wildlife, and all procedures were approved by the Institutional Animal Care and Use Committee of the University of Central Oklahoma. Conflict of Interest The authors declare no conflict of interest. Literature Cited Amo, L., Lopez, P. & Martın, J. 2007: Refuge use: a conflict between avoiding predation and using mass in lizards. Physiol. Behav. 22, 334 343. Baird, T. A. 2008: A growth cost of experimentally induced conspicuous coloration in first-year collared lizard males. Behav. Ecol. 19, 589 593. Baird, T. A. 2013a: Social life on the rocks: behavioral diversity and sexual selection in collared lizards. In: Reptiles in Research: Investigations of Ecology, Physiology, and Behavior From Desert to Sea. (Lutterschmidt, W. I., ed). Nova Biomedical Press, New York, NY, pp. 213 245. Baird, T. A. 2013b: Lizards and other reptiles as model systems for the study of aggressive contest behaviour. In: Animal Contests. (Hardy, I. C. W. & Briffa, M., eds). Cambridge Univ. Press, Cambridge, pp. 258 286. Baird, T. A. & Sloan, C. L. 2003: Interpopulation variation in the social organization of female collared lizards, Crotaphytus collaris. Ethology 109, 879 894. Baird, T. A., Sloan, C. L. & Timanus, D. K. 2001: Intra-and inter-seasonal variation in the socio-spatial behavior of 42 Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH

J. R. York & T. A. Baird Antipredator Tail Displays in a Lizard adult male collared lizards, Crotaphytus collaris (Reptilia, Crotaphytidae). Ethology 107,15 32. Baird, T. A., Timanus, D. K. & Sloan, C. L. 2003: Intra- and intersexual variation in social behavior: effects of ontogeny, phenotype, resources, and season. In: Lizard Social Behavior. (Fox, S. F., McCoy, J. K. & Baird, T. A., eds). Johns Hopkins Univ. Press, Baltimore, MD, pp. 7 46. Bateman, P. W. & Fleming, P. A. 2009: To cut a long tail short: a review of lizard caudal autotomy studies carried out over the last 20 years. J. Zool. 277, 1 14. Bock, W. J. 1958: Preadaptation and multiple evolutionary pathways. Evolution 13, 194 211. Braun, C. A., Baird, T. A. & LeBeau, J. K. 2010: Influence of substrate temperature and directness of approach on the escape responses of juvenile collared lizards. Herpetologica 66, 418 424. Broom, M. & Ruxton, G. D. 2005: You can run, or you can hide: optimal strategies for cryptic prey. Behav. Ecol. 16, 534 540. Burger, J. & Gochfeld, M. 1981: Discrimination of the threat of direct versus tangential approach to the nest by incubating herring and great black-backed gulls. J. Comp. Physiol. Psychol. 95, 676 684. Caro, T. M. 1995: Pursuit-deterrence revisited. Trends Ecol. Evol. 10, 500 503. Caro, T. M. 2005: Antipredator Defences in Birds and Mammals, 1st edn. Univ. Chicago Press, Chicago, IL. Cohen, J. 1992: A power primer. Psychol. Bull. 112, 155. Colegrave, N. & Ruxton, G. D. 2003: Confidence intervals are a more useful complement to nonsignificant tests than are power calculations. Behav. Ecol. 14, 446 447. Cooper, W. E. Jr 1998: Reactive and anticipatory display to deflect predatory attack to an autotomous lizard tail. Can. J. Zool. 76, 1507 1510. Cooper, W. E. Jr 2003: Effect of risk on aspects of escape behavior by a lizard, Holbrookia propinqua, in relation to optimal escape theory. Ethology 109, 617 626. Cooper, W. E. Jr 2008: Visual monitoring of predators: occurrence, cost and benefit for escape. Anim. Behav. 76, 1365 1372. Cooper, W. E. Jr & Vitt, L. J. 2010: Blue tails and autotomy: enhancement of predation avoidance in juvenile skinks. Ethology 70, 265 276. Cooper, W. E. Jr, Perez-Mellado, V., Baird, T. A., Caldwell, J. P. & Vitt, L. J. 2004: Pursuit deterrent signalling by the Bonaire whiptail lizard Cnemidophorus murinus. Behaviour 141, 297 311. Endler, J. A. & Basolo, A. L. 1998: Sensory ecology, receiver biases and sexual selection. Trends Ecol. Evol. 13, 415 420. Font, E., Carazo, P., Perez, G. & Kramer, M. 2012: Predator-elicited foot shakes in wall lizards (Podarcis muralis): evidence for a pursuit-deterrent function. J. Comp. Psychol. 126,87 96. Fox, S. F. & Baird, T. A. 1992: The dear enemy phenomenon in the collared lizard, Crotaphytus collaris, with a cautionary note on experimental methodology. Anim. Behav. 44, 780 782. Gould, S. J. & Vrba, E. S. 1982: Exaptation a missing term in the science of form. Paleobiology 8, 4 15. Greene, H. W. 1988: Antipredator mechanisms in reptiles. In: Biology of the Reptilia, Vol. 16. (Gans, C. & Huey, R. B., eds). John Wiley & Sons, New York, NY, pp. 1 152. Hasson, O., Hibbard, R. & Ceballos, G. 1989: The pursuit deterrent function of tail-wagging in the zebra-tailed lizard (Callisaurus draconoides). Can. J. Zool. 67, 1203 1209. Hill, R. I. & Vaca, J. F. 2004: Differential wing strength in Pierella butterflies (Nymphalidae, Satyrinae) supports the deflection hypothesis. Biotropica 36, 362 370. Husak, J. F. 2006a: Does speed help you survive? A test with collared lizards of different ages. Funct. Ecol. 20, 174 179. Husak, J. F. 2006b: Does survival depend on how fast you can run or how fast you do run? Funct. Ecol. 20, 1080 1086. Husak, J. F. & Fox, S. F. 2006: Field use of maximal sprint speed by collared lizards (Crotaphytus collaris): compensation and sexual selection. Evolution 60, 1888 1895. Johnson, M. A. & Wade, J. 2010: Behavioural display systems across nine Anolis lizard species: sexual dimorphisms in structure and function. Proc. R. Soc. Lond. B Biol. Sci. 277, 1711 1719. Langkilde, T., Schwarzkopf, L. & Alford, R. A. 2004: The function of tail displays in male rainbow skinks (Carlia jarnoldae). J. Herpetol. 37, 328 335. Martın, J. & Lopez, P. 1999: An experimental test of the costs of antipredatory refuge use in the wall lizard, Podarcis muralis. Oikos 84, 499 505. Nakagawa, S. & Cuthill, I. C. 2007: Effect size, confidence interval and statistical significance: a practical guide for biologists. Biol. Rev. (Camb). 82, 591 605. Perez-Tris, J., Di az, J. A. & Telleri a, J. L. 2004: Loss of body mass under predation risk: cost of antipredatory behaviour or adaptive fit-for-escape? Anim. Behav. 67, 511 521. R Development Core Team. 2013: R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.rproject.org/. Ruxton, G. D., Sherratt, T. N. & Speed, M. P. 2004: Avoiding Attack: The Evolutionary Ecology of Crypsis, Warning Signals and Mimicry, 1st edn. Oxford Univ. Press, Oxford, UK. Scarratt, A. M. & Godin, J. G. J. 1992: Foraging and antipredator decisions in the hermit crab Pagurus Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH 43

Antipredator Tail Displays in a Lizard J. R. York & T. A. Baird acadianus (Benedict). J. Exp. Mar. Biol. Ecol. 156, 225 238. Stankowich, T. & Blumstein, D. T. 2005: Fear in animals: a meta-analysis and review of risk assessment. Proc. R. Soc. Lond. B Biol. Sci. 272, 2627 2634. Telemeco, R. S., Baird, T. A. & Shine, R. 2011: Tail waving in a lizard (Bassiana duperreyi) functions to deflect attacks rather than as a pursuit-deterrent signal. Anim. Behav. 82, 369 375. Ydenberg, R. C. & Dill, L. M. 1986: The economics of fleeing from predators. Adv. Study Behav. 16, 229 249. York, J. R. & Baird, T. A. 2015: Testing the adaptive significance of sex-specific mating tactics in collared lizards (Crotaphytus collaris). Biol. J. Linn. Soc. 115, 423 436. York, J. R., Baird, T. A. & Haynie, M. L. 2014: Unexpected high fitness payoff of subordinate social tactics in male collared lizards. Anim. Behav. 91, 17 25. 44 Ethology 122 (2016) 37 44 2015 Blackwell Verlag GmbH