Alternative reproductive modes of Atlantic forest frogs

J Ethol (2012) 30:331–336 DOI 10.1007/s10164-011-0322-9 SHORT COMMUNICATION Alternative reproductive modes of Atlantic forest frogs Luı́s Felipe Toledo • Michel V. Garey • Thais R. N. Costa • Ricardo Lourenço-de-Moraes Marı́lia T. Hartmann • Célio F. B. Haddad • Received: 23 May 2011 / Accepted: 30 November 2011 / Published online: 30 December 2011 Ó Japan Ethological Society and Springer 2011 Abstract Diversity in reproductive modes is well known in amphibians, mainly among anurans, which are characterized by a diversity in breeding biology that exceeds that of any other tetrapod. Currently, 39 reproductive modes are recognized among anurans and some species display more than one mode. The breeding biology of some Brazilian Atlantic forest anurans was investigated for this study. We L. F. Toledo (&) Museu de Zoologia ‘‘Prof. Adão José Cardoso’’, Universidade Estadual de Campinas (UNICAMP-IB), Rua Albert Einstein s/n, Campinas, São Paulo CEP 13083-863, Brazil e-mail: toledolf2@yahoo.com M. V. Garey
T. R. N. Costa Programa de Pós-Graduação em Ecologia e Conservação, Universidade Federal do Paraná. Setor de Ciências Biológicas, Caixa Postal 19031, Curitiba, Paraná CEP 81531-980, Brazil e-mail: michelgarey@gmail.com Present Address: M. V. Garey Departamento de Zoologia e Botânica, Universidade Estadual Paulista, Rua Cristóvão Colombo 2265, São José do Rio Preto, São Paulo CEP 15054-000, Brazil R. Lourenço-de-Moraes Programa de Pós-Graduação em Ecologia e Conservação da Biodiversidade, Universidade Estadual de Santa Cruz, Rodovia Ilhéus-Itabuna, km 16, Ilhéus, Bahia CEP 45662-000, Brazil M. T. Hartmann Universidade Federal do Pampa, Campus de São Gabriel, Av. Antonio Mercado 1357, São Gabriel, Rio Grande do Sul 97300-000, Brazil C. F. B. Haddad Departamento de Zoologia, I.B, Universidade Estadual Paulista, UNESP, Caixa Postal 199, Rio Claro, São Paulo CEP 13506-900, Brazil e-mail: haddad1000@gmail.com observed unreported reproductive modes for six species, variability in the reproductive modes of individuals of the same species (whereby some individuals of a given population displayed unusual reproductive modes when the physical conditions of the breeding site were suboptimal), and variations within the modes. These observations suggest possible evolutionary steps for the reproductive modes. Anuran breeding biology seems to be more diverse than previously reported, and a character matrix could be constructed to describe the total range of variation of the anuran reproductive modes. Keywords Reproduction
Amphibian
Anura
Behavioral variation
Egg-deposition site Introduction Animal reproduction is highly diverse and complex (Leonard 2010). The existence of alternative reproductive strategies may have evolved due to trait selection in order to maximize fitness, avoid predation, and avoid competition; or this evolution of reproductive traits can be related to specific environmental conditions (e.g., Magnusson and Hero 1991; Haddad and Prado 2005; Taborsky et al. 2008; Taborsky and Brockmann 2010). Therefore, in order to clarify the evolution of reproductive traits we must first describe and categorize observed trait variations in a comparative framework. Accordingly, some herpetologists (Salthe and Duellman 1973; Haddad and Prado 2005) have defined some criteria to recognize and distinguish different reproductive modes (RMs) in amphibians; namely, differences in development (the presence or absence of larval stages), egg-laying sites, larval development sites, presence or absence of parental 123 332 care, and type of parental care. These traits can vary between species, within populations of the same species, or even in the same individual, which may display behavioral plasticity (e.g., Touchon and Warkentin 2008). This documented variation suggests that the class Amphibia stands as the tetrapod group with the highest number (39) of RMs (Haddad and Prado 2005). Some RMs are commonly observed across anuran taxa, such as RM 1—eggs laid in lentic bodies of water in which exotrophic tadpoles develop (sensu Haddad and Prado 2005)—a RM that is widespread throughout the clade Anura. In contrast, some RMs are specific to one species or a species group. For example, RMs 15 and 16, in which eggs embedded in the dorsum of aquatic female either hatch into exotrophic tadpoles (RM 15) or into froglets (RM 16), occur only in Pipidae (Haddad and Prado 2005). The similarity of RMs in a given taxonomic group may involve only one character, whereas other characters vary. For example, all species of Leptodactylus (Leptodactylidae) deposit their eggs in foam nests. However, species of Leptodactylus may create foam nests on the water surface, in constructed basins, or in constructed subterranean chambers (Haddad and Prado 2005). Therefore, despite some degree of conservatism, RMs vary interspecifically. Behavior also varies intraspecifically in some taxa. For example, Physalaemus spiniger (Leiuperidae) has at least three different RMs (Haddad and Pombal 1998; Haddad and Prado 2005) and Dendropsophus ebraccatus and Hypsiboas pardalis (Hylidae) have been reported to have at least two RMs (Touchon and Warkentin 2008; Moura et al. 2011). Such variation may be frequent in the Atlantic forest (AF), as it houses more than 530 species of anurans (L. F. Toledo and C. F. B. Haddad, unpublished data), representing about 8% of the total number of anurans known (Frost 2011), many of which are endemic to this biome and are endangered (Haddad et al. 2008). Collectively, this anuran fauna has an unexpected diversity of RMs, with about two-thirds of the known RMs being represented (Haddad and Prado 2005). Therefore, based on our field studies in the AF during the past few years, we present novel observations that may contribute to an understanding of the evolution of RMs. We suggest that we must be more flexible in our classification of reproductive traits to account for such great variations observed in nature. Materials and methods Field expeditions were made in different areas (see below) both during the rainy [from November to March in southern and southeastern Brazil (summer), and from May to August in northeastern Brazil (winter)] and dry [from May to August in southern and southeastern Brazil (winter) and 123 J Ethol (2012) 30:331–336 from September to March in northeastern Brazil (summer)] seasons of the year. There was no effort to create standardization between sampling sites and observations were occasional. Samplings begun in December 1992 and finished in July 2010. Samplings did not extend through all these years at the same site, i.e., sites were surveyed from 1 to 3 consecutive years. During field samplings frogs were found by active search using headlamps. Six anuran species were observed in the field during their reproductive seasons at four sites within the AF: Scinax littoralis and Physalaemus spiniger were studied between September 2006 and November 2008 at Reserva Natural Salto Morato, municipality of Guaraqueçaba, state of Paraná in southern Brazil (RNSM; 25°090 S, 48°160 –48°200 W, 30 m above sea level). Dendropsophus haddadi was studied in July 2010 in Reserva Particular do Patrimônio Natural (RPPN) Capitão, municipality of Itacaré, state of Bahia in northeastern Brazil (14°190 S, 39°040 W, 126 m above sea level). Bokermannohyla sp. (aff. circumdata) and Rhinella ornata were studied between January 2000 and November 2002, in the Parque Estadual da Serra do Mar, Núcleo Picinguaba, municipality of Ubatuba, state of São Paulo in southeastern Brazil (23°230 S, 44°500 W, sea level). Hypsiboas faber was studied between December 1992 and December 1993 on the border of the municipalities of Ribeirão Branco and Apiaı́, state of São Paulo in southeastern Brazil (24°130 S, 48°460 W, 800 m above sea level). Results Bufonidae Rhinella ornata Males call during the rainy season on the ground near lentic and lotic bodies of water. Amplectic pairs deposit eggs in the water along the edges of ponds (RM 1) and rivulets (RM 2). One gelatinous egg string contained 2589 eggs. Hylidae Bokermannohyla sp. (aff. circumdata) During the rainy season males call perched 2 m above the ground along the margins of rivulets and swamps in forested areas. Of the eight spawnings observed in January 2001, one clutch contained 155 eggs; clutches were deposited in small, muddy basins constructed by males (Ivan Sazima, personal communication) in swamps (RM 4). We found one egg mass deposited in a rivulet (RM 2) near the calling site of a male observed on the previous night in August 2002. J Ethol (2012) 30:331–336 333 Fig. 1 Gelatinous egg mass of Dendropsophus haddadi in its first 4 days (a–d, respectively) Dendropsophus haddadi Males call perched along the margins of temporary, lentic bodies of water in the rainy season. We observed an amplectic pair, which laid a gelatinous egg mass on a leaf of a shrub 1.5 m above the water surface (Fig. 1a). We monitored this egg mass for 4 days until the tadpoles dropped into the water. Initially, the egg mass was globular, but gradually it attenuated into an elongated string (Fig. 1a–d). On the 4th day the embryos were active inside the eggs and suddenly nearly the entire string of eggs (21 out of 25 eggs) fell into the water. Tadpoles immediately began to swim and dispersed in the water. This species has RM 24. Hypsiboas faber Males call from the banks of semi-permanent bodies of water during the rainy season. Usually they call near mud nests that they have excavated from a few hours or even days before they begin calling. Water percolates from the adjacent body of water into the muddy-walled nests. More than 90% (45 out of 48 observations) of the clutches were deposited as a film of eggs floating on the surface of the water in the clay nests (Fig. 2a)—RM 4. However, sometimes eggs were deposited outside the nest. If muddy banks are not available, a male may construct a nest amidst the aquatic vegetation by pushing the plants and creating walls of plant matter (2 out of 48 observations). The eggs are deposited inside the nest as a surface film (Fig. 2b). This seems to be a variant of RM 4, given that the constructed nest is in the pond, not at its edge. At sites lacking muddy banks and vegetation, males do not construct a nest and the eggs are deposited directly on the surface of the water (Fig. 2c)—RM 1 (1 out of 48 observations). Some males do not construct nests even when muddy banks are available; instead, they use natural depressions near the body of water and the eggs are deposited as a surface film (Fig. 2d) (2 out of 48 observations). The latter seems to be another variant of RM 4 because the male does not construct a nest. One clutch had approximately 3000 eggs. The mean diameter of the eggs was 2.12 ± 0.28 mm. Scinax littoralis During the rainy season, males call along the margins of temporary or permanent lentic bodies of water in the forest. Most of the time, individuals deposit their eggs directly on the surface of the pond (RM 1). However, in September 2006, an amplectic pair was observed perched in bushy vegetation. After 2 h, the pair moved 50 cm above their perch to an epiphytic bromeliad 138 cm from the ground. The pair remained mostly still in the base of the bromeliad axil for 30 min. After that, the couple moved to two other leaves. The frogs deposited 53 eggs in the water accumulated in the bromeliad leaf axil (Fig. 2e; RM 6). During the same night, two other egg masses were found in axils of bromeliads located 2 and 3 m above the ground near a temporary pond. These clutches had 46 and 58 eggs, respectively. The mean egg diameter was 1.32 ± 0.08 mm (range 1.21–1.49 mm, n = 20). Leiuperidae Physalaemus spiniger During the summer several foam nests were observed on the water surface in the margins of temporary ponds. In addition, another four foam nests were found in natural 123 334 J Ethol (2012) 30:331–336 Fig. 2 Egg clutches of Hypsiboas faber laid in a nest constructed of mud (a), nest constructed of vegetation (b), directly on the pond surface (c), and in a natural depression near the pond (d); a gelatinous egg mass of Scinax littoralis laid in the bromeliad axil (e), and foam nest of Physalaemus spiniger in a depression covered with grass near temporary ponds (f) depressions (i.e., not constructed by the frogs) 1–5 m from temporary ponds or swampy areas. These depressions were shallow and contained a small amount of water and they were partially covered with grass (Fig 2f). The foam nests contained 17–155 eggs. the surface of temporary ponds (RM 11), in terrestrial bromeliad axils (RM 14), and in humid places near temporary ponds (RM 28) (Haddad and Pombal 1998), but not in partially grass-covered basins near temporary ponds (a variation of RM 28). Scinax littoralis was reported to deposit eggs exclusively in temporary ponds (RM 1) (Pombal and Gordo 1991), not in bromeliads (RM 6). Bokermannohyla sp. (aff. circumdata) is an undescribed species for which we report the first reproductive data, indicating that the frogs deposit their eggs either in the flowing waters of rivulets (RM 2) or in water accumulated in muddy basins that they construct (RM 4). Variability in Discussion Our observations suggest that Atlantic rainforest anurans are more variable in their RMs than previously thought. For example, Physalaemus spiniger was known to lay eggs on 123 J Ethol (2012) 30:331–336 reproductive mode (RM) has been reported for R. ornata and H. faber (Haddad et al. 2008), although here we describe novel variations for RM 4 of H. faber. Haddad et al. (2008) reported 22 species characterized by more than one RM. The same authors (Haddad et al. 2008) indicated 12 species (4 families) that varied between RMs 1 and 2, one hylid species that varied between RMs 1 and 4, six leiuperid species that have two modes (RMs 11 and 28), and one leiuperid species that has three modes (RMs 11, 14, and 28; see also Cruz and Peixoto 1985; Haddad and Pombal 1998). Recently, another leiuperid species was reported to share the latter three RMs (Pupin et al. 2010), and two species of Dendropsophus have RMs 1 and 24 (Miranda et al. 2008; Touchon and Warkentin 2008). Thus, reproductive variability, in terms of variation of RMs, seems to be common among Neotropical anurans and may be correlated with specific physical environmental factors, such as the prevalence of shade (Touchon and Warkentin 2008) and the absence of mud along pond edges (present study: in the case of H. faber). At least four RMs occur in the genus Scinax in the AF— RM 1 (e.g., many species of the S. ruber clade and some of the S. catharinae clade); RM 2 (e.g., some species of the S. catharinae clade); RM 6 (species of the S. perpusillus group); and RM 11 (exclusive to S. rizibilis) (Haddad and Prado 2005; Toledo and Haddad 2005). Scinax littoralis is placed in the S. catharinae group (Faivovich 2002), in which RM 6 is not known to occur; RM 6 is known to occur only in the S. perpusillus clade. Therefore, our observation of occasional variation in the RM of S. littoralis may reflect incipient diversification of RMs in Scinax. Thus, an individual that usually engages in RM 1 may also occasionally display RM 6 (plasticity not observed here), and this trait may become fixed in the species. This change from an aquatic to a terrestrial environment could be related to selective pressure exerted by predators, as has been suggested for other anurans (e.g., Heyer 1969; Magnusson and Hero 1991), or to loss of appropriate aquatic habitat, as observed for H. faber in the present study. The novel variation of egg-deposition sites in P. spiniger supports previous observations of high reproductive variability in this species (Haddad and Pombal 1998; Haddad and Prado 2005), and may represent a transitional step from the presumably generalized and widespread ancestral mode of depositing eggs on the surfaces of ponds to laying eggs in bromeliads or on the humid forest floor. Likewise, H. faber exhibits at least two variants of RMs 4 and 1 that suggest a pattern of behavioral specialization. From RM 1, the most likely ancestral mode of modern anurans (Haddad and Prado 2005), H. faber has evolved the abilities to construct nests and to lay eggs by the side of ponds. These features may represent intermediate steps in the evolution of RM 4 from RM 1. 335 The variation observed in R. ornata is common in other species (as in the other 12 species reported by Haddad et al. 2008), indicating that species that lay eggs in still water may also be able to deposit eggs in flowing water, within acceptable bounds of water speed and turbulence. Therefore, the selection of egg-laying sites in anurans with regard to water systems may represent a continuum of possibilities, rather than a strict choice as to lentic or lotic bodies of water. Prior to the present study, the egg-laying site of D. haddadi and its RM were unknown. However, it was suggested that the site would be similar to that of closely related species that deposit their eggs on leaves above the water (Bastos and Pombal 1996). We have validated this suggestion and described variation in the shape and development of the egg mass that we think is novel among anurans. A similar egg string was illustrated for D. bokermanni (Crump 2009: Fig 1.2), but we cannot confirm that it is the same strategy as that observed for D. haddadi. Laying eggs on leaves, instead of directly in the water, may prevent egg predation (e.g., Heyer 1969; Magnusson and Hero 1991). However, hatchlings may be preyed upon just when they drop into the water. Buchacher (1993), for example, reported individuals of Pipa arrabali feeding on falling hatchling Phyllomedusa bicolor tadpoles. Therefore, hatching in groups and simultaneously may reduce predation on hatchlings, as predicted by the satiation hypothesis of Begon et al. (2006). As evidenced by the present study, we can anticipate discovering new variations in anuran reproductive behavior and ecology. As we learn more, we may find it useful to expand our definitions of RMs to include a combination of characters and character states that may represent continuous, rather than categorical variables (Haddad and Prado 2005). Thus, it may be possible to construct a matrix with several reproductive traits and include additional characters and character states, such as the shape of the gelatinous egg mass, more accurate and restrictive descriptions of egglaying sites, or the continuous variation of egg-laying sites, as discussed above. With such an approach, we think it would be possible to reorganize the current 39 RMs into more categories that more accurately reflect the behavior of anuran taxa and may contribute to a better interpretation of the evolution of anuran reproductive biology. Acknowledgments Two anonymous reviewers provided insightful suggestions in earlier versions of the manuscript. Rogério P. Bastos, José J. Pombal Jr., Tomaz Fumio Takeuchi, and Diogo B. Provete helped with the fieldwork. Ivan Sazima provided unpublished information on Bokermannohyla sp. (aff. circumdata). David Rodriguez helped with English corrections. MVG acknowledges Fundação O Boticário de Proteção à Natureza (convênio FBPN-UFPR 623) for financial support and FAPESP for a fellowship (process: 2008/505751). IBAMA provided collecting permits (no. 107/06-RAN). MTH acknowledges Capes and CNPq for fellowships. TRNC thanks Capes 123 336 for a scholarship. LFT thanks FAPESP (2008/50325-5) for Grants and a scholarship (2008/52847-9). RLM thanks Capes for a scholarship. CFBH thanks FAPESP and CNPq for financial support. References Bastos RP, Pombal JP Jr (1996) A new species Hyla (Anura: Hylidae) from eastern Brazil. Amphibia Reptilia 17:325–331 Begon M, Townsend CR, Harper JL (2006) Ecology: from individuals to ecosystems. Blackwell, Oxford Buchacher CO (1993) Field studies on the small Surinam toad, Pipa arrabali, near Manaus, Brazil. Amphibia Reptilia 14:59–69 Crump ML (2009) Amphibian diversity and life history. In: Dodd CK Jr (ed) Amphibian ecology and conservation: a handbook of techniques. Oxford University Press, New York, pp 3–20 Cruz CAG, Peixoto OL (1985) About the spawn of Physalaemus in an unusual site (Amphibia, Anura, Leptodactylidae). Arq Univ Rural Rio Janeiro 8:103–105 Faivovich J (2002) A cladistic analysis of Scinax (Anura: Hylidae). Cladistics 18:367–393 Frost DR (2011) Amphibian species of the world: an online reference. V 5.5. Available at: http://research.amnh.org/herpetology/ amphibia/index.php. American Museum of Natural History, New York. Accessed on 31 January 2011 Haddad CFB, Pombal JP Jr (1998) Redescription of Physalaemus spiniger (Anura: Leptodactylidae) and description of two new reproductive modes. J Herpetol 32:557–565 Haddad CBF, Prado CPA (2005) Reproductive modes in frogs and their unexpected diversity in the Atlantic forest of Brazil. Bioscience 55:207–217 Haddad CFB, Toledo LF, Prado CPA (2008) Atlantic forest amphibians. Editora Neotropica, São Paulo Heyer WR (1969) The adaptive ecology of the species groups of the genus Leptodactylus (Amphibia, Leptodactylidae). Evolution 23:421–428 Leonard JL (2010) Introduction: celebrating and understanding reproductive diversity. In: Leonard JL, Córdoba-Aguilar A 123 J Ethol (2012) 30:331–336 (eds) The evolution of primary sexual characters in animals. Oxford University Press, New York, pp 1–5 Magnusson WE, Hero JM (1991) Predation and the evolution of complex oviposition behaviour in Amazon rainforest frogs. Oecologia 86:310–318 Miranda DB, Garey MV, Monteiro-Filho ELA, Hartmann MT (2008) Visual signaling and reproductive biology of Dendropsophus werneri (Anura: Hylidae) in an area of Atlantic forest in the state of Paraná, Brazil. Pap Av Zool 48:335–343 Moura MR, Motta AP, Feio RN (2011) An unusual reproductive mode in Hypsiboas (Anura: Hylidae). Zoologia 28:142–144 Pombal JP Jr, Gordo M (1991) Two new species of Hyla from the Atlantic forest in the state of São Paulo (Amphibia, Anura). Mem Inst Butantan 53:135–144 Pupin NC, Gasparini JL, Bastos RP, Haddad CFB, Prado CPA (2010) Reproductive biology of an endemic Physalaemus of the Brazilian Atlantic forest, and the trade-off between clutch and egg size in terrestrial breeders of the P. signifer group. Herpetol J 20:147–156 Salthe SN, Duellman WE (1973) Quantitative constraints associated with reproductive mode in anurans. In: Vial JL (ed) Evolutionary biology of the anurans. University of Missouri Press, Columbia, pp 229–249 Taborsky M, Brockmann HJ (2010) Alternative reproductive tactics and life history phenotypes. In: Kappeler P (ed) Animal behaviour: evolution and mechanisms. Springer, Heidelberg, pp 537–586 Taborsky M, Oliveira RF, Brockmann HJ (2008) The evolution of alternative reproductive tactics: concepts and questions. In: Oliveira RF, Taborsky M, Brockmann HJ (eds) Alternative reproductive tactics: an integrative approach. Cambridge University Press, Cambridge, pp 1–21 Toledo LF, Haddad CFB (2005) Reproductive biology of Scinax fuscomarginatus (Anura, Hylidae) in south-eastern Brazil. J Nat Hist 39:3029–3037 Touchon JC, Warkentin KM (2008) Reproductive mode plasticity: aquatic and terrestrial oviposition in a treefrog. Proc Natl Acad Sci USA 105:7495–7499