Herpetology Notes, volume 3: 301-304 (2010) (published online on 22 November 2010)

Predation on Allobates femoralis (Boulenger 1884; Anura: Aromobatidae) by the colubrid snake Xenopholis scalaris (Wucherer 1861)

Max Ringler1*, Eva Ursprung1, Walter Hödl1

Subsumed under the vernacular name “poison frogs”, and Vitt, 2009). Despite intensive research on various the superfamily Dendrobatoidea contains the “non- aspects of dendrobatoid biology, little information is poisonous poison frogs” in the family Aromobatidae available on natural predators of this taxon (cf. Darst and and the “true poison frogs” in the family Dendrobatidae Cummings, 2006; Saporito et al., 2007; Cooper, Caldwell (Grant et al., 2006). While the latter regularly feature and Vitt, 2009; Noonan and Comeault, 2009). Allobates bright, aposematic colouration and potent skin toxins, femoralis is a semi-cryptic pan-Amazonian aromobatid members of the Aromobatidae are usually cryptically frog (Amézquita et al., 2009) and possesses, at the most, coloured and rely on camouflage and rapid escape only traces of skin toxins of the potent alkaloid classes behaviour as anti-predator measures (Cooper, Caldwell know from dendrobatids (Daly, Myers and Whittaker, 1987). However, the species was found to be a Batesian mimic of poisonous species of the dendrobatid genus Epipedobates (Darst and Cummings, 2006), including 1 Department of Evolutionary Biology, University of Vienna E. hahneli, which is syntopic with A. femoralis at the Althanstraße, 14, A-1090 Vienna, Austria; email: [email protected] observation site in French Guiana (Lescure and Marty, * corresponding author. 2000; Born and Gaucher, 2001). Additionally, the skin

Figure 1. Adult Allobates femoralis with attached HDF reflector. 302 Max Ringler et al.

The focus individual, a male A. femoralis with a snout-urostyle length of 27.05 mm and a weight of 1.7 g, was used in field experiments to test the suitability of the harmonic-direction-finding (HDF) telemetry- technique (Pellet et al., 2006) to study male tadpole transport and female mating commutation in this highly territorial (Ringler, Ursprung and Hödl, 2009) species. The trials took place in tropical lowland rainforest near the research station “Saut Pararé” in the nature reserve “Les Nouragues”, French Guiana (4.04° N, 52.68° W, WGS84). After some other individuals had been observed in captivity in order to optimise reflector attachment, on 28 February 2008 three individuals, including the focal male, were released inside their respective territories with attached reflectors to investigate their behaviour under natural conditions. The reflectors consisted of a Schottky diode that was soldered between two antennas made of 40 µm steel- strands forming a 6 cm by 12 cm T-shaped dipole with Figure 2. Adult Xenopholis scalaris coiled under leaves. the braze points covered in shrinkable tubing. To attach the reflectors to the frogs we used waistbands (Fig. 1), excretions of A. femoralis are likely to make the frog made from latex-free condoms to prevent skin irritation unpalatable to some predators due to their bitter taste (Gutleb et al., 2001). Together with the waistband, the (pers. obs). reflectors had a total mass of less than 0.05 g, which is

Figure 3. Partly digested Allobates femoralis recovered from partly skeletonised Xenopholis scalaris. Predation of Xenopholis scalaris upon Allobates femoralis 303 less than 5% of the body mass of adult A. femoralis. Acknowledgements. Fieldwork was supported by the Austrian The focal male was recovered with a RECCO HDF Science Fund grant FWF-18811 (PI Walter Hödl). We are several times from 28 - 29 February and showed especially grateful to Nicolas Perrin from the Department of Ecology and Evolution at the UNIL, Lausanne, Switzerland for apparently normal behaviour during that time, including the loan of the RECCO HDF. Permissions were provided by the constant advertisement calls and regular movement and CNRS and fieldwork was conducted in compliance with current escape behaviour. On 1 March, with the first detection French and EU law. on that day at 1100 hours, we unambiguously retrieved the reflector’s signal from an adult (aprox. 300 mm in total length) Xenopholis scalaris, which we discovered References typically coiled in its protective position (Zug, Vitt and Caldwell, 2007) under some leaves (Fig. 2), still inside Amézquita, A., Lima, A.P., Jehle, R., Castellanos, L., Ramos, O., the territory of the focal male. To observe the further Crawford, A., Gasser, H., Hödl, W. (2009): Calls, colours, sha- pe, and genes: a multi-trait approach to the study of geographic behaviour of the snake, we took it into captivity in a variation in the Amazonian frog Allobates femoralis. Biol. J. netted cage. During the night the snake died and was Linn. Soc. 98: 826-838. already partly skeletonised by ants when we discovered Born, M.G., Gaucher, P. (2001): Distribution and life histories of it the next morning. We opened the body cavity of the amphibians and reptiles. In: Bongers F, Charles-Dominique P, snake and recovered the partly digested frog, which Forget P, Théry M (Eds.): Nouragues. Dynamics and plant-ani- apparently had been swallowed head first and whose mal interactions in a neotropical rainforest. Kluwer Academic ventral pattern was still recognizable for individual Publishers, Dordrecht: 167-184. identification (Fig. 3). No apparent internal injuries Cooper, W.E., Caldwell, J.P., Vitt, L.J. (2009): Conspicuousness and vestigial escape behaviour by two dendrobatid frogs, Den- from the reflector were visible and its antennae were drobates auratus and Oophaga pumilio. Behaviour 146: 325- folded smoothly in the intestinal tract of the snake. 349. Hence we cannot be certain whether the death of the Daly, J.W., Myers, C.W., Whittaker, N. (1987): Further classifica- snake was caused by toxic skin excretions from the frog, tion of skin alkaloids from neotropical poison frogs (Dendro- direct effects of the swallowed reflector, or if it was batidae), with a general survey of toxic/noxious substances in attacked and killed by the ants that covered the corpse the amphibia. Toxicon 25: 1023-1095. in the morning. Darst, C.R., Cummings, M.E. (2006): Predator learning favours mimicry of a less-toxic model in poison frogs. Nature 440: Xenopholis scalaris is a small (300-350 mm in total 208-211. length) leaf-litter snake with an equally pan-Amazonian Grant, T., Frost, D.R., Caldwell, J.P., Gagliardo, R., Haddad, distribution as A. femoralis where it inhabits primary C.F.B., Kok, P.J.R., Means, D.B., Noonan, B.P., Schargel, and old secondary rainforest in humid zones. It features W.E., Wheeler, W.C. (2006): Phylogenetic systematics of a brick-red brown body with a dark longitudinal central Dart-Poison Frogs and their relatives (Amphibia: Athesphata- line running along the body which is accompanied by nura: Dendrobatidae). Bull. Am. Mus. Nat. Hist. 299. Pp: 262. perpendicular running, dark triangular shaped patterns Gutleb, A.C., Bronkhorst, M., van den Berg, J.H.J., Murk, A.J. (Fig. 2). The species is known to feed on frogs, and (2001): Latex laboratory-gloves: an unexpected pitfall in am- phibian toxicity assays with tadpoles. Environ. Toxicol. Phar- given its nocturnal activity (Starace, 1998), it is highly macol. 10: 119-121. likely that the prey frog was taken during the night Jehle, R., Arntzen, J.W. (2000): Post-breeding migrations of new- when A. femoralis, like almost all dendrobatoids, is not ts (Triturus cristatus and T. marmoratus) with contrasting eco- active. Snake predation on telemetered amphibians has logical requirements. J. Zool. 251: 297-306. been reported previously by Spieler and Linsenmair Lescure, J., Marty, C. (2000): Atlas des amphibiens de Guyane. (1998) and Jehle and Arntzen (2000). However, as we Paris, Muséum national d’histoire naturelle. did not witness the actual predation event, we can only Noonan, B.P., Comeault, A.A. (2009): The role of predator selec- tion on polymorphic aposematic poison frogs. Biology Letters speculate whether the attached reflector rendered the 5: 51-54. frog more conspicuous and made it more susceptible to Pellet, J., Rechsteiner, L., Skrivervik, A.K., Zürcher, J.-F., Perrin, fall prey to the snake, or if the reflector hindered the N. (2006): Use of the harmonic direction finder to study the frog in escaping its predator. terrestrial habitats of the European tree frog (Hyla arborea). Amphibia-Reptilia 27: 138-142. Ringler, M., Ursprung, E., Hödl, W. (2009): Site fidelity and pat- terns of short- and long-term movement in the brilliant-thighed poison frog Allobates femoralis (Aromobatidae). Behav. Ecol. Sociobiol. 63: 1281-1293. 304 Max Ringler et al.

Saporito, R.A., Zuercher Rachel, Roberts, M., Gerow, K.G., Don- Starace, F. (1998): Guide des serpents et amphisbènes de Guyan- nelly, M.A. (2007): Experimental evidence for aposematism in ne française. Guadeloupe, Guyane, Ibis Rouge Editions. the Dendrobatid poison frog Oophaga pumilio. Copeia 2007: Zug, G.R., Vitt, L.J., Caldwell, J.P. (2007): Herpetology. An in- 1006-1011. troductory biology of amphibians and reptiles. San Diego, Ca- Spieler, M., Linsenmair, K.E. (1998): Migration patterns and di- lifornia, Academic Press. urnal use of shelter in a ranid frog of a West African savannah: a telemetric study. Amphibia-Reptilia 19: 43-64.

Accepted by Miguel Vences