Factors Affecting the Population Dynamics of Two Toads (Anura: Bufonidae) in a Seasonal Neotropical Savanna Mariana M. Vasconcellos 1 and Guarino R. Colli 1 We studied the population dynamics of Bufo schneideri and B. rubescens during three years in the Cerrado biome of central Brazil. Using mark–recapture models, we estimated demographic parameters and evaluated the effects of sex and climate upon them. We used model selection, based on the Akaike’s information criterion (AIC c ), to test hypotheses of variation in demographic parameters. The adult sex ratio was male-biased for both species, and breeding took place in the dry season for B. rubescens and in the transition from dry to wet season for B. schneideri. There was no difference in recapture probability between sexes, but there was significant variation among months. Recapture probability was associated with rainfall and temperature of the previous month in B. rubescens, but had constant values for the same months in different years in B. schneideri, showing a cyclic seasonal pattern. Survival estimates were constant in time and no significant difference was found between sexes in both species. Annual survival was 0.35 for females and 0.37 for males of B. schneideri, and 0.34 for males and 0.38 for females of B. rubescens. We found no evidence for temporary emigration in both populations, but the presence of transients was significant. For both species, adult population size was moderate, with larger values during the breeding season. The similar demography of B. rubescens and B. schneideri illustrates the high ecological and morphological conservatism reported in the genus Bufo. T HE global amphibian decline phenomenon (Beebee and Griffiths, 2005), associated with population decreases and extinctions in many regions of the world (Houlahan et al., 2000), has enormously increased the interest, need, and importance of amphibian monitoring studies. This interest is especially focused on the Neotropical region, where enigmatic declines have been more promi- nent (Stuart et al., 2004), with particular attention to Latin America, which bears the highest amphibian diversity but where amphibian population studies are meager (Young et al., 2001; Lips et al., 2005). These declines have intensively affected species of Bufonidae (true toads), one of the most threatened group of amphibians in the New World, with 55% of its species being extinct or threatened (Young et al., 2004). The estimation of basic demographic parameters and the identification of their determinants are essential to improve the understanding of population dynamics and life history of toads (Schmidt et al., 2002). Many recent advances have been made in population modeling analysis, using capture– recapture models to accurately estimate demographic parameters, such as population size and separate estimates of survival and capture probabilities, providing important tools for the identification of population trends (Lebreton et al., 1992; Schmidt, 2003). The focus of these analyses has also been directed to model selection, used to test hypoth- eses of biological interest and to identify important sources of variation in life history features (Lebreton et al., 1992; Nichols, 1992; Johnson and Omland, 2004). Amphibians with complex life cycles can be differently affected by several factors acting on different life stages, such as egg, larval, juvenile, or adult (Wilbur, 1980). Nevertheless, Biek et al. (2002) demonstrated that survival in post- metamorphic stages of anurans, especially juvenile and adult stages, have a major impact on population growth rates, playing a special role in population dynamics. Zug and Zug (1979) estimated that, at the most, 0.5% of the individuals of Bufo marinus survive from egg to sexual maturity in their natural habitats. Indeed, many biotic and abiotic factors may influence demographic parameters of the terrestrial stage of toads. Climate factors, mainly rainfall and temperature, strongly influence amphibian phenology (Aichinger, 1987; Beebee, 1995), promoting seasonal varia- tion in breeding activity and demographic parameters. Individual features, such as age, sex, size, and genotype, can also affect anuran survival probabilities (Wood et al., 1998; Holenweg Peter, 2001). Bufo schneideri is a common, large toad broadly distributed in open areas of South America (Pramuk, 2006); neverthe- less, its biology is poorly known. Breeding generally takes place from August to October, during the transition of dry to wet season (Rossa-Feres and Jim, 1994; Moreira and Barreto, 1997), with a maximum time of larval development of two months (Barreto and Moreira, 1996). Likewise, little is known on the biology of B. rubescens, which is apparently endemic to the Cerrado biome (Colli et al., 2002). Breeding apparently occurs during the dry season, between May and September (Eterovick and Sazima, 2004). These two species are closely related, pertaining to the B. marinus group (Pramuk, 2006), with females often being larger than males (Merchan et al., 2001; Eterovick and Sazima, 2004). Herein, based on three years of mark–recapture studies in the Cerrado biome of central Brazil, we investigate the popula- tion dynamics of B. schneideri and B. rubescens, estimating demographic parameters and modeling the effects of sex and climate variations. MATERIALS AND METHODS Study site and field methods.—The study was conducted in Estac ¸a ˜o Ecolo ´gica de A ´ guas Emendadas (ESECAE), a protect- ed area in Distrito Federal, Brazil (15u359220S, 47u419470W). The area is in the core region of Cerrado (altitude of 900 m), a savannic biome considered a biodiversity hotspot for conservation priority (Myers et al., 2000). The Cerrado originally covered approximately 2 million km 2 , or 22% of the Brazilian territory, and has suffered intense deforesta- tion (for a review on the Cerrado biome, see Oliveira and Marquis, 2002). Climate is strongly predictable and season- 1 Departamento de Zoologia, Universidade de Brası ´lia, 70910-900 Brası ´lia, Distrito Federal, Brazil; E-mail: (GRC) grcolli@unb.br. Send reprint requests to GRC. Submitted: 17 April 2007. Accepted: 6 November 2008. Associate Editor: G. Haenel. F 2009 by the American Society of Ichthyologists and Herpetologists DOI: 10.1643/CE-07-099 Copeia 2009, No. 2, 266–276