ISSN 1827-9635 (print) © Firenze University Press ISSN 1827-9643 (online) www.fupress.com/ah Acta Herpetologica 9(1): 99-102, 2014 DOI: 10.13128/Acta_Herpetol-13606 Observations on the use of tarantula burrows by the anurans Leptodactylus bufonius (Leptodactylidae) and Rhinella major (Bufonidae) in the Dry Chaco ecoregion of Bolivia Christopher M. Schalk1, Marco Sezano2 1Ecology and Evolutionary Biology Program, Department of Wildlife and Fisheries Sciences, and Biodiversity Research and Teaching Collections, Texas A&M University, College Station, Texas, USA. Corresponding author. E-mail: cschalk@tamu.edu 2Sección de Herpetología, Museo de Historia Natural Noel Kempff Mercado, Santa Cruz de la Sierra, Bolivia Submitted on 2013,21st November; revised on 2014, 31st January; accepted on 2014, 2nd February Abstract. Some species of anurans have been observed utilizing burrows of other animals, such as rodents and taran- tulas. Here we report the observations of two anuran species, Leptodactylus bufonius and Rhinella major, utilizing the burrows of tarantulas (Acanthoscurria sp.; Family Theraphosidae) in the dry Chaco ecoregion of Bolivia. Both species of anurans never co-occurred with tarantulas in the burrows and used burrows that were wider in diameter and closer to breeding ponds as compared to the total available tarantula burrows in the area. These burrows may serve as ref- uges from predators, especially for conspicuous, calling males. Keywords. Acanthoscurria, amphibian, Neotropics, predation, refuge, spider, Theraphosidae, tropical dry forest. A number of animals are known to utilize cavities created by other animals, such as the many cavity-nest- ing birds that require tree cavities created by other ani- mals for reproduction (Scott et al., 1977). Previous work has documented anurans using burrows of other ani- mals including rodents, tortoises, and tarantulas (Gen- try and Smith, 1968; Cocroft and Hambler, 1989; Witz et al., 1991; Dundee et al., 2012; Schalk, 2012). While spiders are known predators of anurans at both the lar- val (Schulze and Jansen, 2010) and post-metamorphic (Toledo, 2005) life stages, surveys of tarantula (Fam- ily Theraphosidae) burrows in both North America and South America have found that species of frogs in the family Microhylidae actually take refuge within an occu- pied tarantula burrow (Blair, 1936; Cocroft and Hambler, 1989; Dundee et al., 2012). A study of the skin chemistry of Gastrophryne carolinensis (Family Microhylidae) dem- onstrated that they produce toxic skin secretions mak- ing them unpalatable to spiders and thus allow them to coexist with the tarantulas within their burrows (Garton and Mushinsky, 1979). Although previous studies have focused on anuran use of burrows still occupied by the tarantula, anurans and other taxa can also utilize aban- doned burrows (Witz et al., 1991). During a 50-day survey in the Dry Chaco ecoregion of Bolivia, we observed two species of anurans, Lepto- dactylus bufonius (Family Leptodactylidae) and Rhinella major (Family Bufonidae) utilizing tarantula (Acanthoscu- rria sp., Family Theraphosidae) burrows. To our knowl- edge, this is the first instance of anuran species outside the family Microhylidae using tarantula burrows. Also, while anurans in the Gran Chaco have been observed utilizing the burrows of the Vizcacha (Lagostomus maximus; Fam- ily Chinchillidae) (Cei, 1980; Schalk, 2012), this is the first documentation of anurans in the Gran Chaco ecoregion utilizing tarantula burrows. The objectives of this study were to document if these anurans 1) overlapped tempo- rally with tarantulas in their occupancy of a burrow, and 2) utilized a non-random subset of the available burrows based on microhabitat characteristics of the burrows. 100 C.M. Schalk, M. Sezano The study was conducted in an area of the Boliv- ian Gran Chaco within the vicinity of Yande Yari (18° 41’ 30.516” S, 62° 18’ 6.9474” W), a park guard camp in the Kaa-Iya of the Gran Chaco National Park, Cordil- lera Province, Santa Cruz Department, Bolivia. The cli- mate of the Chaco is seasonal, having a wet, hot summer (November–March), and a dry, cool winter (April–Octo- ber). Average rainfall and temperature for this area are 513 mm and 24.6 ºC, respectively (Navarro and Mal- donado, 2002). The area is characterized by scrubby, short trees (e.g., Schinopsis lorentzii and Aspidosperma quebracho-blanco), while shrubs (Capparis sp., Acacia sp.), bromeliads, and cactii, (e.g., Opuntia sp., Cleisto- cactus baumannii and Eriocereus guelichii) are common understory plants (Navarro and Maldonado, 2002). As part of an effort to document the reptiles and amphibians in the area, we conducted nightly time-con- strained visual encounter surveys from 9 February 2012 to 29 March 2012 (Schalk et al., in press). Our visual encounter surveys were conducted between 20:00 h and 00:00 h along one of four trails at the camp. Each night we surveyed one of the four available trails, which was ran- domly chosen each night. While conducting these visual encounter surveys, we also conducted a two-step survey process of the occupants of the tarantula burrows found along the trail. The first step consisted of identifying the burrow to be that of a tarantula’s by confirming it as an occupant. When we encountered a burrow, we shined our flashlight down a burrow to determine if it was occupied. When a burrow was occupied by a tarantula, we measured its diameter to the nearest mm using a ruler, and meas- ured the distance to the nearest source of water to the nearest cm using a 50 m measuring tape. The burrow was then marked using bright flagging tied to a nearby plant or bush to assist in locating the burrow upon subsequent visits. The second step of the process consisted of revisit- ing the marked tarantula burrows at a later date during the nightly survey to determine if they contained an anu- ran as an occupant. If the burrow was found to be occu- pied by an anuran on a subsequent visit, we then meas- ured the same variables previously mentioned. We meas- ured burrow diameter to examine whether these anurans were excluded from utilizing burrows of smaller diameters due to their body size. We measured distance to a pond because we hypothesized that frogs might utilize burrows closer to breeding ponds as refuges from predators. Giv- en that calling males are more conspicuous to predators (Ryan et al., 1981; Schalk and Morales, 2012), we hypoth- esized that those unoccupied tarantula burrows that were in close proximity to breeding sites may offer some form of protection to the frogs if they were approached by a predator. Inherent in this hypothesis is that the frogs exhibit a preference to utilize tarantula burrows closer to ponds; therefore we assumed that the likelihood of a bur- row being abandoned by a tarantula and therefore, the availability of abandoned tarantula burrows was inde- pendent relative to the distance to a pond. Since the bur- rows occupied by frogs were previously documented to be occupied by tarantulas from the first step in the survey process, we did not include the frog-occupied burrows in our analysis of the tarantula burrows as it would be tem- poral pseudoreplication (Hurlbert, 1984). The data failed to meet the assumptions for parametric t-tests, thus we conducted a nonparametric Mann-Whitney U-test com- paring the burrows occupied only by tarantulas to those occupied by anurans. We found a total of 54 occupied tarantula burrows in the study area. Of the 54 burrows, 46 of the burrows only contained tarantulas as occupants, while eight of the 54 burrows were found to be utilized by anurans of two species when we re-surveyed the burrows. We did not observe frogs and tarantulas utilizing the same bur- row simultaneously. Of the eight frog observations, six were male L. bufonius, while the other two were a male and female R. major. All of the male L. bufonius were calling from within the burrow or at the edge of the bur- row and when approached, entered the burrow for ref- uge. The male R. major was not observed calling. Dur- ing our surveys, we did witness a predation attempt in which a tarantula tried to capture a female L. bufonius. The burrows utilized by the frogs were significantly larger (median = 31.9 mm) in their diameter than those bur- rows in which we only found tarantulas (median = 23.7 mm; Fig. 1A; Mann-Whitney, U = 86, P = 0.01757). The burrows used by the frogs were also significantly closer (median = 3.5 m) to a source of water as compared to the burrows used only by tarantulas (median = 12.7 m; Fig. 1B; Mann-Whitney, U = 75, P < 0.01). All of the male L. bufonius used burrows less than four meters from the edge of a pond, whereas the male and female R. major were using burrows 5.4 m and 18.7 m from a pond, respectively. Burrows often provide an amenable environment that offers shelter from harsh abiotic conditions or refuge from predators (Cocroft and Hambler, 1989; Witz et al., 1991). In the Chaco ecoregion, several anurans had been documented as being burrow associates of the Vizcacha (L. maximus); the known burrow associates included L. bufonius (Cei, 1980), as well as Leptodactylus laticeps (Cei, 1980), and Physalaemus biligonigerus (Schalk, 2012). The frogs utilized a subset of size classes of the total available burrows used by tarantulas in the area. Specifi- cally, they used burrows which were larger in diameter and closer to a water source. Both L. bufonius and R. 101Use of tarantula burrows by anurans major are moderately sized anurans (mean SVL = 53 mm and 58.8 mm, respectively) and may be unable to utilize the smaller diameter burrows as a result. All six male L. bufonius using the abandoned taran- tula burrows were observed calling in or close to the tarantula burrow entrance, and all sought refuge inside the burrow when approached. Predation rates are often skewed towards male frogs at a breeding site as males are more conspicuous than females (Ryan et al., 1981; Lodé, 1996). These abandoned tarantula burrows may offer some form of protection to the more conspicuous males. In this region, documented predators of calling males include invertebrates (Schalk, 2010), foxes (Schalk and Morales, 2012), and other anurans (Schalk and Mon- taña, 2011). The breeding biology of L. bufonius may also permit it to utilize these burrows several meters from the pond’s edge. The males of L. bufonius construct under- ground nest chambers over a meter from the pond’s edge where the eggs are deposited and the nest is sealed with mud; the tadpoles can persist inside until the nest is flooded by heavy rains and the tadpoles enter the nearby pond (Cei, 1980; Reading and Jofré, 2003). Rhinella major call from the pond’s edge (Schalk and Morales, 2012) and deposit eggs directly in the water (Cei, 1980), therefore they may be unable to utilize these burrows that are sev- eral meters away from the pond as refuges during breed- ing bouts. The feeding experiments in Cocroft and Hambler’s (1989) demonstrate that leptodactylid and bufonid frogs were palatable to the tarantula Xenesthis immanis. Dur- ing our surveys, we observed an unsuccessful predation attempt on a female L. bufonius by a tarantula; the taran- tula ambushed the L. bufonius, but it was able to escape quickly. We also never observed either species of frog and tarantulas co-occurring in the same burrow at the same time, suggesting that antagonistic interactions in the form of a predator-prey relationship may be occur- ring. However, without conducting feeding experiments (sensu Cocroft and Hambler, 1989; Dundee et al., 2012), we cannot comment on the palatability of L. bufonius or R. major to the tarantulas in the area. ACKNOWLEDGEMENTS This research was conducted as an agreement between the Biodiversity Research and Teaching Collections at Texas A&M University and the Kaa-Iya of the Gran Chaco National Park. We thank R.L. Cuellar and K. Rivero for permit support and the park guards C. Socoré, G. Depita, G. Castro, and J. Alupi for their assistance in the field. We also thank R. Bertani for identifying the tarantula. C.G. Montaña, A. Schulze, and the Langerhans Lab at NCSU provided constructive comments on the manuscript. Support for CMS was provided by the National Science Foundation’s Graduate Research Fellowship Program and the Applied Biodiversity Science NSF-IGERT Program at Texas A&M University (NSF-IGERT Award # 0654377). This is publication number 1467 of the Biodiversity Research and Teaching Collections at Texas A&M University. REFERENCES Blair, W.F. (1936): A note on the ecology of Microhyla oli- vacea. Copeia 1936: 115. Cei, J.M. (1980): Amphibians of Argentina. Mon. Zool. Ital. Mono. 2: 1–609. Cocroft, R.B., Hambler, K. (1989): Observations on a commensal relationship of the microhylid frog Chias- mocleis ventrimaculata and the burrowing theraphosid Fig. 1. Boxplots of the A) diameter of the burrows, and B) the dis- tance to a pond used by tarantulas and frogs during the survey period. An asterisk (*) indicates P < .05, and a double asterisk (**) indicates P < .01. Outliers are depicted by an open circle symbol. 102 C.M. Schalk, M. Sezano spider Xenesthis immanis in southeastern Peru. Bio- tropica 21: 2–8. Dundee, H.A., Shillington, C., Yeary, C.M. (2012): Inter- actions between tarantulas (Aphonopelma hentzi) and narrow-mouthed toads (Gastrophryne olivacea): sup- port for a symbiotic relationship. Tulane Studies Zool. Bot. 32: 31–38. Garton, J.D., Mushinsky, H.R. (1979): Integumentary tox- icity and unpalatability as an antipredator mechanism in the narrow mouthed toad, Gastrophryne carolinen- sis. Can. J. Zool. 57: 1965–1973. Gentry, J.B., Smith, H.M. (1968): Food habits and burrow associates of Peromyscus polionotus. J. Mammol. 49: 562–565. Hurlbert, S.H. (1984): Pseudoreplication and the design of ecological field experiments. Ecol. Mono. 54: 187– 211. Lodé, T. (1996): Polecat predation on frogs and toads at breeding sites in western France. Ethol. Ecol. Evol. 8: 115–124. Navarro, G., Maldonado, M. (2002): Geografía Ecológica de Bolivia: Vegetación y Ambientes Acuáticos. Santa Cruz de la Sierra. Centro de Ecológia Difusión Simon I Patiño. 719 pp. Reading, C.J., Jofré, G.M. (2003): Reproduction in the nest building vizcacheras frog Leptodactylus bufonius in central Argentina. Amphibia Reptilia 24: 415-428. Ryan, M. J., Tuttle, M.D., Taft, L.K. (1981): The costs and benefits of frog chorusing behavior. Behav. Ecol. Soc. 8: 273–278. Schalk, C.M. (2010): Physalaemus biligonigerus (NCN) Predation. Herpetol. Rev. 41: 202. Schalk, C.M. (2012): Physalaemus biligonigerus Burrow use. Herpetol. Rev. 43: 123–124. Schalk, C.M., Montaña, C.G. (2011): Ceratophrys cranwel- li (Cranwell’s Horned Frog) Diet. Herpetol. Rev. 42: 409–410. Schalk, C.M., Morales, F. (2012): Predation of a Rhinella major (Anura: Bufonidae) by a Pampas Fox (Lyca- lopex gymnocercus) in the Bolivian Gran Chaco. Her- petol. Notes 5: 369–370. Schalk, C.M., Sezano, M., Cuellar, R.L. In press. Inven- tory of the reptiles and amphibians from a locality in the Kaa-Iya of the Gran Chaco National Park, Bolivia. Kempffiana. Schulze, A., Jansen, M. (2010): A tadpole of Trachyceph- alus venulosus (Anura: Hylidae) as prey for a fishing spider (Araneae: Pisauridae) in the Bolivian Chiqui- tano Dry Forest. Herpetol. Notes 3: 297–298. Scott, V.E., Evans, K.E., Patton, D.R., Stone, C.P. (1977): Cavity nesting birds of North American forests. Agri- cultural Handbook 511, U.S. Dept. Agriculture, Wash- ington, D.C., USA. Toledo, L.F. (2005): Predation of juvenile and adult anu- rans by invertebrates: current knowledge and perspec- tives. Herpetol. Rev. 36: 395–400. Witz, B.W., Wilson, D.S., Palmer, M.D. (1991): Distribu- tion of Gopherus polyphemus and its vertebrate symbi- onts in three burrow categories. Amer. Mid. Nat. 126: 152–158. Acta Herpetologica Vol. 8, n. 2 - December 2013 Firenze University Press Journal of the Societas Herpetologica Italica ACTA HERPETOLOGICA