Tails of enticement: caudal luring by an ambushforaging

Transcription

1 Functional Ecology 2008, 22, doi: /j x Tails of enticement: caudal luring by an ambushforaging snake (Acanthophis praelongus, Blackwell Publishing Ltd Elapidae) M. Hagman, B. L. Phillips and R. Shine* School of Biological Sciences A08, University of Sydney, NSW 2006, Australia Summary 1. Ambush foragers that attract prey via luring provide an opportunity to examine how a predator s behaviour influences its dietary composition. 2. Australian death adders (Acanthophis praelongus, Elapidae) are heavy-bodied ambush foragers with broad diets; the snake s modified tail-tip is waved to attract prey. Female adders have shorter tails than males, but longer terminal spines. 3. We videotaped captive snakes interacting with potential prey items (lizards and frogs) to document which prey types elicit luring, and which respond by approaching the lure. To clarify prey responses, we controlled lure size and colour by attaching snake tails (removed from dead adders) to a machine that waved the tail-tip in a manner similar to a live snake. 4. Individual adders differed in luring behaviour, and the type of tail-tip movement (undulatory vs. straightline) influenced rates and duration of luring bouts. 5. Lure movement was essential to attract lizards, and small lures were more effective than larger ones; the greater effectiveness of small lures may explain why caudal luring tends to be more common in juvenile snakes than in larger conspecifics. 6. Death adders lured more vigorously to lizards than to frogs, and lizards were more likely to approach the lure. Thus, luring in death adders mostly enables these snakes to capture lizards; frogs (also an important dietary component in the field) must be caught another way. 7. An ambush predator s overall dietary composition, as well as ontogenetic changes in that composition, thus depend upon both lure characteristics and prey responses. Key-words: dietary composition, foraging mode, frogs, lizards, prey types, sit-and-wait predator Functional Ecology (2008) xx, Introduction Predators display a complex repertoire of behaviours in the course of locating, capturing and handling prey, and intuition suggests that many of these behaviours are adaptations to capture specific types of prey. For example, some predators detect and follow faint chemical traces left by potential prey species, ignoring similar cues from other taxa (e.g. Greenlees, Webb & Shine 2005). Others select specific foraging sites that maximize their opportunities to detect and capture specific prey types (Shine & Sun 2002). Yet others exhibit abilities or weapons well-suited to specific types of prey: for example, cheetahs can outrun fleet-footed gazelles, and the venoms of predators often work most effectively to immobilize the prey taxa upon which they specialize (Endler 1986; Greene 1988). Despite such matches between form and function, however, empirical evidence on causation is weak; for the vast majority *Correspondence author. rics@bio.usyd.edu.au Present address: Department of Zoology, Stockholm University, Stockholm, Sweden. of predator behaviours we must rely on intuition to judge how a specific phenotypic trait enhances the predator s ability to capture some specific type of prey. Ambush predators that lure their prey provide an unusually robust opportunity to document causal links between predator tactics and dietary composition (Reiserer 2002). Predators of many phylogenetic lineages (e.g. mantids, spiders, fishes, frogs, turtles, lizards, snakes, cats) capture their prey from ambush, remaining hidden until they can launch a foraging strike. In several of these lineages, the predators possess modified body parts that function to lure prey close enough to seize. Well-known examples include the luminous organs of deep-sea anglerfishes (Gudger 1945) and carnivorous fireflies (Lloyd 1965), the modified tongues of alligator snapping turtles (Drummond & Gordon 1979), tails of snakes (Henderson 1970; Greene & Campbell 1972; Heatwole & Davison 1976; Jackson & Martin 1980) and toes of frogs and toads (Murphy 1976; Radcliffe et al. 1986; Hagman & Shine 2008). Such systems facilitate empirical analyses of the relationship between predator attributes and choice of prey. That is, we can control or manipulate the luring stimulus to examine how specific 2008 The Authors. Journal compilation 2008 British Ecological Society

2 Caudal luring by snakes 1135 aspects of the luring system influence responses (and thus, availability to the predator) of different kinds of prey (Reiserer 2002; Hagman & Shine 2008). We have adopted this approach to explore the role of caudal luring in the trophic ecology of an ambush-foraging Australian snake. Most species of Australian venomous snakes belong to the proteroglyphous (front-fanged) lineage Elapidae, and are slender-bodied species that rely upon active searching to find their prey (Cogger 1992). However, a few taxa depart from this morphology and behaviour, and instead are heavyset ambush foragers (Shine 1980; Reed & Shine 2002). Death adders (Acanthophis), the most extreme example of this morphotype, are widely distributed throughout continental Australia and islands to the north, and feed on a wide variety of vertebrate prey (mostly frogs, lizards and small mammals: Shine 1980; Cogger 1992). Acanthophis species are behaviourally, ecologically and morphologically convergent with ambushforaging snakes (e.g. vipers, rattlesnakes) from other parts of the world (Shine 1980). Death adders have relatively short laterally compressed tails with a distinctive terminal spine and typically, a small patch of lighter colour on the ventral surface of the tail (Cogger 1992). Caudal luring has been reported many times in this lineage, especially in juvenile snakes (Carpenter 1978; Chiszar et al. 1990). Materials and methods STUDY SPECIES AND AREA We studied northern death adders (Acanthophis praelongus hawkei: Wuster et al. 2005) on the Adelaide River floodplain 60 km east of Darwin, Northern Territory (12 39 S, E). The area lies within the wet-dry tropics, and is hot year-round but with highly seasonal precipitation (Shine & Brown 2008). Monsoonal rains inundate lowlying areas for a few months each year, with the landscape reverting to dry savanna woodland as the floods recede (Shine & Brown 2008). Death adders in this area range from 230 to 880 mm snout-vent length (SVL) and weigh from 18 to 486 g (B.L. Phillips, unpublished data). The snakes take a wide variety of prey items, with juveniles mostly feeding on frogs and lizards whereas adults (especially females, which grow larger than males) take mammals and occasional birds (Webb, Shine & Christian 2005). METHODS To quantify overall attributes of lures, we measured sizes of caudal lures of Acanthophis praelongus, Ramsay 1877 from our study area (based on preserved specimens in the collection of the Northern Territory Museum). We also captured death adders, frogs and lizards from the Adelaide River floodplain and quantified their behavioural characteristics in the laboratory. We placed death adders (n = 65; SVL ranges mm) in individual glass aquaria ( mm floor area) lined with paper, and given water ad libitum. At dusk we added a single prey item, either a frog (Limnodynastes convexiusculus, Macleay 1878 [ g], Litoria dahlii, Boulenger 1896 [ g], or L. nasuta, Gray 1842 [ g]; total frogs, N = 50) or a lizard (Carlia gracilis, Storr, 1974 [ g] or Glaphyromorphus douglasi, Storr 1967 [ g]; total lizards, N = 39). We videotaped the resultant Fig. 1. Caudal luring by death adder (Acanthophis praelongus). The Figure shows six sequential frames from a video-clip of horizontal caudal undulations; the video-clip is available on the journal s website as Appendix S1 in Supplementary Material. encounter for one hour, and examined the video to score whether or not the adder displayed caudal luring; and if so, whether or not the potential prey item approached the lure (i.e. moved straight towards the lure in an obvious attempt to investigate it). We videotaped an additional 73 trials of 11 captive snakes (5 10 trials per snake) to quantify kinematics of the lure s movements (Fig. 1). From these latter videos, we scored the form of horizontal tail-tip movement during luring (undulatory vs. straightline), the distance moved by the tail-tip per luring cycle, the length of the tail segment that moved during the luring bout, the frequency of cycles (sideways movements/ sec), and the total duration of each luring bout. To clarify responses elicited by the caudal lure, we also exposed potential prey items to an artificial lure (the tail of a death adder, connected to a machine that moved it in a natural fashion to mimic the rates and distances of movement by the tail-tip during luring by captive snakes). Each trial ran for 5 min, during which lure movement was kept constant. The machine was modified from one used to mimic toe-waving (prey-luring) behaviour by cane toads (see Hagman & Shine 2008 for details). A 12-volt electrical motor rotated (on a horizontal axis) a circular plate of aluminium to which we affixed (slightly off-centre) a flat metallic arm ( mm) holding the tail of a death adder affixed so that it projected horizontally from the end of the arm. The motor was connected to a dimmer, which enabled us to adjust the rotation speed of the aluminium plate so that the lure

3 1136 M. Hagman et al. moved in a fashion resembling that of a real snake (analyses of videorecordings showed that the tip of the artificial lure moved 8 10 mm from side to side at a rate of 4 5 sideway movements per second). The tail (removed from a road-killed snake) protruded through a small hole on the floor at one end of a plastic bin ( mm) housing either a frog (n = 576) or a lizard (n = 96). To assure statistical independence we used each lizard or a frog only once, in one trial. We videotaped the prey s response in terms of whether or not it attacked the lure. Tails removed from freshly-killed snakes remained supple and hence usable for a few days postmortem; in total, we used 12 tails over the course of our experiments. To clarify how attributes of the lure affect its effectiveness, we used tails from dead adders of a range of body sizes (small 250 mm SVL; medium = mm SVL; large 880 mm SVL) and painted some tails to modify their colour. Results MORPHOLOGY OF THE LURE In 79 museum specimens (SVL range mm), 32 female adders averaged 541 mm SVL whereas 47 males averaged 509 mm SVL (F 1,77 = 0 61, P = 0 44). Males had longer tails relative to SVL (slopes homogeneous, intercept F 1,75 = 15 72, P < ). Terminal spines of females averaged larger than those of males in absolute terms (means of 3 0 vs. 2 6 mm, F 1,67 = 2 88, P < 0 04) and also relative to tail length (slopes homogeneous, intercept F 1,66 = 6 74, P < 0 02) or SVL (F 1,66 = 4 49, P < 0 04). The size of the ventral patch of colour on the tail did not differ between the sexes either in absolute terms (anova, F 1,67 = 0 11, P = 0 74) or relative to tail length (ancova, F 1,65 = 0 97, P = 0 33). KINEMATICS OF LURING anovas with snake identification number as the factor revealed significant individual variation in luring behaviour among the 11 captive adders. The tail-tip always moved in a broadly horizontal plane (Fig. 1), but in some cases it moved straight from side-to-side whereas in other cases the tail-tip undulated as it travelled horizontally. There were significant differences among individual snakes in the relative frequencies of the two types of luring bouts (ranging from 100% of one to 100% of the other; comparing snakes, χ 2 = 19 91, d.f. = 10, P < 0 035). Snakes also differed significantly in the mean length of tail that was moved in luring (F 10,62 = 2 71, P < 0 01), the frequency of cycles per second (F 10,62 = 4 16, P < 0 003), and the total duration of luring bouts (F 10,62 = 6 36, P < 0 001), but the average distance moved by the tail-tip during a luring cycle did not vary significantly among individuals (F 10,62 = 1 24, P = 0 28). The significant individual variation in luring behaviour was not associated with the sex of the snake, for any of the traits we measured (using nested anova with individual ID nested within sex, all P > 0 07), nor with the animal s body length (regression analyses comparing SVL to mean values for behavioural traits, all P > 0 20). Death adders that moved their tail-tips further in each cycle did so more rapidly (mean distance moved by tail-tip vs. mean duration of luring bout, N = 11 snakes, r = 0 62, P < 0 05), and this source of variation was associated with whether the tail-tip was moved in a straightline vs. undulatory fashion (using mean values per snake, snakes that mostly exhibited undulatory bouts took longer per bout: % undulatory bouts vs. mean duration, n = 11, r = 0 63, P < 0 04). These analyses suggest that the type of luring bout (i.e. whether the tail moves in a straightline or undulatory fashion) is an important axis of variation. To examine this question, we analysed the data using nested anova with luring type nested within individual snake ID (9 of the 11 snakes performed both straightline and undulatory luring bouts). The effect of luring type was tested against the among-individual variation not the residual error term. The type of lure movement affected the distance travelled by the tail-tip (straightline displays involved a wider arc and a slower cycling rate, F 1,18 = 20 62, P < and F 1,18 = 6 77, P < 0 02 respectively). Bouts of undulatory cycles lasted longer than those of straightline movements (F 1,18 = 6 41, P < 0 025; see Fig. 2). Snakes also tended to use a greater length of tail in straightline displays than in undulatory displays, but this difference fell short of statistical significance (F 1,18 = 4 39, P = 0 051). WHAT PREY TYPES ELICIT LURING? The species of prey offered to captive death adders affected the probability that the snakes would exhibit luring (log-likelihood ratio from logistic regression, χ 2 = 57 40, d.f. = 4, P < ). Death adders frequently exhibited caudal luring in the presence of lizards (elicited by 17 of 20 Glaphyromorphus, 13 of 19 Carlia, = 30 of 39, = 77% overall) but not frogs (7 of 17 Litoria dahlii, 0 of 16 L. nasuta, 0 of 17 Limnodynastes convexiusculus, = 7 of 50, = 14% overall). If we restrict analysis to trials involving lizards, neither the species of lizard, nor its size relative to the predator s size, influenced the probability that the snake would exhibit caudal luring (lizard species, χ 2 = 1 17, d.f. = 1, P = 0 28; relative prey mass, χ 2 = 2 74, d.f. = 1, P = 0 10). WHAT PREY TYPES RESPOND TO LURING? No frogs approached the luring death adder, but many lizards responded (16 of 17 Glaphyromorphus approached the lure, as did 3 of 13 Carlia; comparing these two species, χ 2 = 19 23, d.f. = 1, P < ). HOW DO ATTRIBUTES OF THE ARTIFICIAL LURE AFFECT PREY RESPONSES? As in the trials using live snakes, lizards were more likely to approach the artificial lure (55 of 96, = 57%) than were frogs (4 of 576, < 1%; χ 2 = , d.f. = 1, P < ). The four frogs that responded comprised two L. nasuta and two C. australis, so we did not detect significant interspecific variation in response frequencies among the four species of frogs tested (χ 2 = 5 57, d.f. = 3, P = 0 13). Lizards were more reactive. Multiple logistic regression analysis on data from lizard trials

4 Caudal luring by snakes 1137 immobile (0 of 24 trials). In trials where we manipulated colour of the lure, the natural colour (brown with yellow patch) attracted prey in 19 of 24 trials (79%), similar to responses elicited by a lure painted all-brown (20 of 24, = 83%) or painted all-yellow (16 of 24, = 67%; comparing responses to the three colours, χ 2 = 2 60, d.f. = 2, P = 0 27). Only one frog (a Cyclorana australis) attacked and seized the artificial lure, but 18 lizards did so. Most attacks came from Glaphyromorphus (16 of 33 vs. 2 of 22 Carlia: χ 2 = 10 25, d.f. = 1, P < 0 002). The lure s colour (χ 2 = 2 09, d.f. = 2, P = 0 35) and size (χ 2 = 0 20, d.f. = 2, P = 0 90) did not influence the probability of attack. Discussion Fig. 2. Relationships between display types and features of caudal luring in captive death adders (Acanthophis praelongus). The Figure shows the relationship between the trajectory of the tail-tip s movement (i.e. either straightline from side-to-side, or undulatory from side-to-side) on (a) distance moved by the tail-tip during each cycle, (b) the frequency of tail-tip displacement cycles, and (c) the total duration of the luring bout. showed that the probability of a lizard approaching the lure differed between the two lizard species tested and was affected by the lure s movement and size, but not by the colour of the lure. Overall, 33 of 48 Glaphyromorphus approached the lure (69%), compared to 22 of 48 Carlia (46%; χ 2 = 11 30, d.f. = 1, P < 0 001). Large lures were approached by lizards in 15 of 32 trials (47%), medium-sized lures in 18 of 32 trials (56%) and small lures in 22 of 32 trials (69%; χ 2 = 7 42, d.f. = 2, P < 0 025). No prey items approached the lure when it was Our results support and extend those of Reiserer s (2002) pioneering study on the interaction between caudal luring snakes and their prey, and provide information on a phylogenetically independent evolution of caudal luring in snakes (in an elapid, vs. Reiserer s viperids). The clear response patterns obtained in our study suggest that luring systems offer excellent models with which to test hypotheses about the ways in which predator tactics and morphology influence the types and numbers of prey captured. To understand communication in a predator prey system, we need information on interactions in relatively natural circumstances (Ford & Burghardt 1993) but to tease apart the role of specific variables, we need to simplify the situation to obtain greater control. Thus, our study included both descriptive components (lure attributes; incidence of luring and prey responses in captive encounters using live snakes) and a manipulative component (the artificial lure). The combination of the two proved useful. For example, the artificial lure enabled us to (i) quantify responses to the lure by a prey type (frogs) that rarely elicited luring (without the artificial lure, we could not determine whether items that did not elicit luring would have responded to it had it occurred); (ii) eliminate potential confounding variables, such as subtle changes in snake behaviour and luring movements stimulated by the prey item s initial responses (such changes might otherwise create artefactual correlations between prey that elicit luring and those that respond to luring); and (iii) decouple attributes such as the size of the lure from confounding issues such as the body size or behaviour of the snake. Because the only stimulus in the artificial lure trials was the lure itself, moved in a consistent fashion, we can be confident that differential responses (such as to lure size) were direct effects of the lure itself rather than responses to unmeasured correlated variables. Differences in prey responses to the artificial lure vs. to live snakes hint that such correlated variables may indeed play a role in prey responses. For example, the two scincid lizard species that we tested differed in their responses to live snakes vs. the artificial lure. Glaphyromorphus douglasi approached live snakes more often than the artificial lure, whereas the reverse was true for Carlia gracilis. The probable reason for that difference was the presence of the death adder (less well-hidden in the laboratory than in leaf-litter in the field).

5 1138 M. Hagman et al. Carlia were warier than Glaphyromorphus (e.g. were less likely to attack either the snake s tail or the artificial lure), and may have detected the snake s presence by visual or chemical cues. Hence, the effectiveness of caudal luring may differ not only among alternative prey species, but in a context-dependent way: some prey types never respond to the lure, others are likely to respond under a wide range of conditions, and yet others respond only when conditions render the snake inconspicuous. In most respects, our data support intuition-based speculations on the function of caudal luring. For example, an immobile lure failed to attract potential prey, confirming the critical role of lure movement in prey attraction. Similarly, luring was elicited only by proximity of prey, supporting the role of this distinctive behaviour in foraging. Third, the moving lure attracted close approach (and sometimes, direct attack) by edible-sized prey, and thus, would have increased the predator s opportunity to seize the item. Hence, the caudal lure is indeed elicited by prey approach, and does enhance opportunities for prey capture. We found considerable variation in the form of the luring display (e.g. in straightness of tail-tip displacement, bout duration, etc.). Part of that variation was among individuals, and part was related to the type (straightline vs. undulatory) of the tail-tip movement. Most snakes displayed both kinds of bouts, and further work could usefully examine the factors influencing display type. We found no clear trends in display attributes with the snake s sex or body size, but studied only 11 animals. Additional work is needed to explore such patterns. For example, does the larger terminal spine on the tails of female adders in some way compensate for their relatively shorter tails? Sex differences in tail length are common in snakes (King 1987), and are accompanied by sexual dichromatism in tail-tip colour in at least one caudal-luring viperid (Bothrops atrox: Burger & Smith 1950). Sex differences in tail length resulting from sexual selection and fecundity selection (King 1987; Shine et al. 1999) may secondarily affect the animal s ability to move its tail in a way that attracts prey. We found that smaller tails were more effective at eliciting approach by potential prey items. Luring thus may work most effectively for small snakes. In keeping with this prediction, caudal luring is more frequently displayed by juveniles than by conspecific adults (Heatwole & Davison 1976; Reiserer 2002). Plausibly, a shift in effectiveness of caudal luring with body size might explain the common ontogenetic dietary shift in prey types in ambush-foraging snakes (Avila-Villegas, Martins & Arnaud 2007; Shine & Wall 2007). However, similar shifts also occur in snake taxa that do not ambush their prey (Shine & Wall 2007), so this cannot be the explanation for all cases. The effectiveness of luring in attracting prey differs not only as a function of the lure s attributes, but also of the prey taxa involved. In the case of death adders, the lure attracted lizards more than frogs (61% vs. 0% approaches to live snakes, 57% vs. < 1% approaches to the artificial lure). A lizard s presence also was likely to elicit luring (77%, vs. 14% for frogs). Thus, death adders initiate caudal luring when they encounter the prey types likely to respond to the lure. The virtual irrelevance of luring to predation on frogs is the most surprising result from our analysis. Palpation of field-collected death adders at our study site yielded 119 prey records: 47 (39%) frogs and 37 (31%) lizards (Webb et al. 2005; J.K. Webb, personal communication). Thus, despite the common assumption that luring plays a central role in foraging success of a caudal-luring species, our data reveal a more complex scenario. Caudal luring attracted one component of the diet, but played no role with the type of prey most frequently consumed. Our results on caudal luring effectiveness for different prey types, and by lures of different sizes, clarify potential mechanisms for ontogenetic shifts in dietary composition. For death adders, the proportion of the diet comprising lizards (the prey type most vulnerable to luring, especially by a small tail) likely would decrease as the snake grew larger, whereas the proportion of frogs (unresponsive to luring, and hence unaffected by lure size) would not. Data from field-captured snakes (J.K. Webb, personal communication) show exactly this ontogenetic shift in the numbers of lizards taken (decreasing from 33 of 75 items from juvenile snakes [44%] to 4 of 44 in adults [9%; χ 2 = 14 20, d.f. = 1, P < 0 001) whereas the incidence of frogs in the diet remains high throughout the snake s life (27 of 75 [36%] in juveniles vs. 20 of 44 [45%] in adults; χ 2 = 0 68, d.f. = 1, P = 0 41). The generalized diet of the death adder must involve a diversity of foraging tactics, of which the most distinctive (caudal luring) is effective only for one component of the diet and a component that is important primarily early in life, when the lure is smallest and thus, most effective. In contrast to our results with A. praelongus, Reiserer (2002) found that American copperheads (Agkistrodon contortrix) and massasauga rattlesnakes (Sistrurus catenatus tergeminus) caudal-lured successfully for small frogs, whereas lizards rarely elicited caudal luring (Reiserer 2002). However, a desert subspecies of massasauga (S. c. edwardsii) lured for lizards (like our death adders) and did not eat frogs. African horned adders (Bitis caudalis) lured successfully for lizards, but not mice (Reiserer 2002). Despite this interspecific divergence, some generalities emerge. For example, caudal-luring is triggered by some prey types and not others; and it is elicited most readily by prey taxa that are most likely to respond to it. Caudal luring has evolved and/or been lost many times within snake phylogeny (Heatwole & Davison 1976; Murphy, Carpenter & Gillingham 1978; Jackson & Martin 1980; Radcliffe, Chiszar & Smith 1980; Sazima 1991; Sazima & Puorto 1993; Leal & Thomas 1994; Tiebout 1997; Simon, Whittaker & Shine 1999; Parellada & Santos 2002) and many snake taxa that caudal-lure exhibit significant geographic, ontogenetic and sex-related variation in dietary composition (Rabatsky & Waterman 2005; Shine & Wall 2005), further enhancing our ability to tease apart causal influences on lure structure and function. There may also be great variation in the specific abiotic (thermal, hydric) conditions that stimulate luring (Rabatsky & Farrell 1996). Caudal luring thus offers an unusually clear example of predator-prey communication, and provides an exciting opportunity to conduct direct experimental tests of hypotheses about the proximate determinants of dietary composition.

6 Caudal luring by snakes 1139 Acknowledgements The Authors thank Michelle Gray for help in the laboratory and Sam Ruggeri for technical assistance. Thomas Madsen, Bea Ujvari, and Matt Greenlees helped us collect death adders, and David Nelson and Georgia Ward-Fear assisted with collecting frogs and lizards. Paul Horner and Dane Trembath (Northern Territory Museum) facilitated our examination of museum specimens. The Australian Research Council funded our work. Permits for the research were provided by the University of Sydney Animal Care and Ethics Committee, and the Parks and Wildlife Commission of the Northern Territory. References Avila-Villegas, H., Martins, M. & Arnaud, G. (2007) Feeding ecology of the endemic rattleless rattlesnake, Crotalus catalinensis, of Santa Catalina Island, Gulf of California, Mexico. Copeia, 2007, Burger, W.L. & Smith, P.W. (1950) The coloration of the tail tip of young Fer-de-Lances: sexual dimorphism rather than adaptive coloration. Science, 112, Carpenter, C.C. (1978) Tail luring in the death adder, Acanthophis antarcticus (Reptilia, Serpentes, Elapidae). Journal of Herpetology, 12, Chiszar, D., Boyer, D., Lee, R., Murphy, J.B. & Radcliffe, C.W. (1990) Caudal luring in the southern death adder, Acanthophis antarcticus. Journal of Herpetology, 24, Cogger, H.G. (1992) Reptiles and Amphibians of Australia, 4th edn. Reed Books, Sydney. Drummond, H. & Gordon, E.R. (1979) Luring in the neonate alligator snapping turtle (Macroclemys temminckii): description and experimental analysis. Journal of Comparative Ethology, 50, Endler, J. (1986) Defense Against Predators. University of Chicago Press, Chicago. Ford, N.B. & Burghardt, G.M. (1993) Perceptual mechanisms and the behavioral ecology of snakes. Snakes: Ecology and Behavior (eds R.A. Siegel & J.T. Collins), pp McGraw-Hill, New York. Greene, H.W. (1988) Antipredator mechanisms in reptiles. Biology of the Reptilia. Volume 16 (eds C. Gans & R.B. Huey), pp Alan R. Liss, New York. Greene, H.W. & Campbell, J.A. (1972) Notes on the use of caudal lures in arboreal green pit vipers. Herpetologica, 28, Greenlees, M., Webb, J.K. & Shine, R. (2005) Led by the blind: bandy bandy snakes Vermicella annulata (Elapidae) follow blindsnake chemical trails. Copeia, 2005, Gudger, E.W. (1945) The anglerfishes, Lophius piscatorius et americanus, use the lure in fishing. American Naturalist, 79, Hagman, M. & Shine, R. (2008) Deceptive digits: the functional significance of toe-waving by cannibalistic cane toads (Chaunus marinus). Animal Behaviour, 75, Heatwole, H. & Davison, E. (1976) A review of caudal luring in snakes with notes on its occurrence in the Saharan sand viper, Cerastes cerastes. Herpetologica, 32, Henderson, R.W. (1970) Caudal luring in a juvenile Russell s viper. Herpetologica, 26, Jackson, J.F. & Martin, D.L. (1980) Caudal luring in the dusky pygmy rattlesnake, Sistrurus miliarius barbouri. Copeia, 1980, King, R.B. (1987) Sexual dimorphism in snake tail length: a test of alternative hypotheses. American Zoologist, 27, 26A. Leal, M. & Thomas, R. (1994) Notes on the feeding behaviour and caudal luring by juvenile Alsophis portoricensis (Serpentes, Colubridae). Journal of Herpetology, 28, Lloyd, J.E. (1965) Aggressive mimicry in Photuris: firefly femmes fatales. Science, 149, Murphy, J.B. (1976) Pedal luring in the leptodactylid frog, Ceratophrys calcarata. Herpetologica, 32, Murphy, J.B., Carpenter, C.C. & Gillingham, J.C. (1978) Caudal luring in the green tree python, Chondropython viridis (Reptilia, Serpentes, Boidae). Journal of Herpetology, 12, Parellada, X. & Santos, X. (2002) Caudal luring in free-ranging adult Vipera latasti. Amphibia-Reptilia, 23, Rabatsky, A.M. & Farrell, T. (1996) The effects of age and light level on foraging posture and frequency of caudal luring in the rattlesnake, Sistrurus miliarius barbouri. Journal of Herpetology, 30, Rabatsky, A.M. & Waterman, J.M. (2005) Ontogenetic shifts and sex difference in caudal luring in the dusky pygmy rattlesnake, Sistrurus miliarius barbouri. Herpetologica, 61, Radcliffe, C.W., Chiszar, D. & Smith, H.B. (1980) Prey-induced caudal movements in Boa constrictor with comments on the evolution of caudal luring. Bulletin of the Maryland Herpetological Society, 16, Radcliffe, C.W., Chiszar, D., Estep, K., Murphy, J.B. & Smith, H.M. (1986) Observations on pedal luring and pedal movements in Leptodactylid frogs. Journal of Herpetology, 20, Reed, R.N. & Shine, R. (2002) Lying in wait for extinction: ecological correlates of conservation status among Australian elapid snakes. Conservation Biology, 16, Reiserer, R.S. (2002) Stimulus control of caudal luring and other feeding responses: a program for research on visual perception in vipers. Biology of the Vipers (eds G.W. Schuett, M. Höggren, M.E. Douglas & H.W. Greene), pp Eagle Mountain Publishing, Utah. Sazima, I. (1991) Caudal luring in two neotropical pitvipers Bothrops jararaca and B. jararacussu. Copeia, 1991, Sazima, I. & Puorto, G. (1993) Feeding technique of juvenile Tropidodryas striaticeps: probable caudal luring in a colubrid snake. Copeia, 1993, Shine, R. & Brown, G.P. (2008) Adapting to the unpredictable: reproductive biology of vertebrates in the Australian wet-dry tropics. Invited review. Philosophical Transactions of the Royal Society Series B, 363, Shine, R. & Sun, L. (2002) Arboreal ambush site selection by pit-vipers (Gloydius shedaoensis). Animal Behaviour, 63, Shine, R. & Wall, M. (2005) Ecological divergence between the sexes in reptiles. Sexual Segregation in Vertebrates (eds P. Neuhaus & K.E. Ruckstuhl), pp Cambridge University Press, Cambridge. Shine, R. (1980) Ecology of the Australian death adder, Acanthophis antarcticus (Elapidae): evidence for convergence with the Viperidae. Herpetologica, 36, Shine, R. & Wall, M. (2007) Why is intraspecific variation in foraging biology more common in snakes than in lizards? Lizard Ecology (eds S.M. Reilly, L.B. McBrayer & D.B. Miles), pp Cambridge University Press, Cambridge. Shine, R., Olsson, M.M., Moore, I.T., Lemaster, M.P. & Mason, R.T. (1999) Why do male snakes have longer tails than females? Proceedings of the Royal Society Series B, 266, Simon, P., Whittaker, R. & Shine, R. (1999) Morelia spilota (Australian carpet python). Caudal luring. Herpetological Review, 30, Tiebout, H.M. III. (1997) Caudal luring by a temperate colubrid snake, Elaphe obsoleta, and its implications for the evolution of the rattle among rattlesnakes. Journal of Herpetology, 31, Webb, J.K., Shine, R. & Christian, K.A. (2005) Intraspecific niche partitioning in a native predator may influence its response to invasion by a toxic prey species. Austral Ecology, 30, Wuster, W., Dumbrell, A.J., Hay, C., Pook, C.E., Williams, D.J. & Fry, B.G. (2005) Snakes across the Strait: trans-torresian phylogeographic relationships in three genera of Australasian snakes (Serpentes: Elapidae: Acanthophis, Oxyuranus, and Pseudechis). Molecular Phylogenetics and Evolution, 34, Received 19 May 2008; accepted 16 July 2008 Handling Editor: Frank Messina Supporting Information Additional Supporting Information may be found in the online version of this article: Appendix S1. Video-clip of death adder caudal luring. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.