Herpetology Notes, volume 7: 347-353 (2014) (published online on 26 May 2014)

Morphometric variation of Pipa pipa (Linnaeus, 1758) (Anura: ) with notes on diet and gonad development

Helena Nery Alves-Pinto1, Vanessa Kruth Verdade2* and Miguel Trefaut Rodrigues1,3

Abstract. Pipa pipa is an emblematic and endemic aquatic from the Amazonian Basin, but much of its biology remains poorly known. In this paper we focused on morphometric variation, diet and gonad examination of P. pipa. We measured 23 morphometric variables in 56 individuals of P. pipa and tested for sexual dimorphism and for geographic variation using Principal Component Analysis. Data on diet and gonad development were studied in one population. The morphometric study showed that forearm diameter, finger and snout length are sexually dimorphic characteristics, all larger in males. The most abundant item found in the species’ diet was Cyclopoida, whereas Heteroptera represented the group with the highest occurrence in stomachs.

Keywords: Pipidae; sexual dimorphism; diet; morphology; reproduction

Introduction and behavioural characteristics (Trueb and Massemin 2000). The reproductive behaviour involves a turnover The frog family Pipidae occurs in Africa and America action, during which the couple swims in vertical circles and includes peculiar aquatic species (Canatella and in the water together while the eggs are released and laid Trueb, 1988; Duellman and Trueb, 1994; Frost et al., on the female back (Rabb and Rabb, 1960; Rabb and 2006). It is monophyletic and one of the most basal Snedigar, 1960; Rabb, 1961). Males are smaller than anuran groups (Canatella and Trueb, 1988; Frost et al., females and have thicker forearms (Zippel, 2006). The 2006, Irisarri et al. 2011). Among the four genera in the female’s vent is rounder and more turgid, especially family (Hymenochirus, Pipa, Pseudhymenochirus and during reproductive periods (Rabb and Snedigar, Xenopus), Pipa is the only genus occurring in Central 1960; Rabb, 1961; Zippel, 2006). However, no other and South America (Dunn, 1948; Canatella and Trueb, characteristics have been investigated regarding sexual 1988; Frost, 2011), from which four out of seven dimorphism and those listed are either subjective or species occur in Brazil - P. arrabali Izecksohn 1976, observed only during reproductive periods. P. snethlageae Müller 1914, P. carvalhoi (Miranda- Aspects of the biology of Pipa pipa other than Ribeiro, 1937), and P. pipa (Trueb and Cannatella, amplexus behaviour have also been neglected and 1986). most information available comes mainly from captive Pipa pipa, the largest species in the genus (Dunn, individuals (Zippel, 2006). Literature on diet of the 1948; Rabb, 1961), has distinct external, osteological species recorded fish as well as anurans of genus Leptodactylus (Deckert, 1917; Duellman, 1978). During foraging, the species uses a combination of inertial suction and forelimb movements, and sometimes prey can be grasped between digits (Carreno and Nishikawa, 1 Universidade de São Paulo, Instituto de Biociências, 2010). Departamento de Zoologia, Rua do Matão, travessa 14, nº Reproductive biology and diet are relevant traits 10, Butantan 05422-970, São Paulo, SP, Brazil, e-mail: of natural history that affect survival, define habitats [email protected] and behaviours (Cuello et al., 2006). The lack of 2 Universidade Federal do ABC, Centro de Ciências Naturais e information on the ecology and biology of species is Humanas, Avenida dos Estados 5001, Bairro Bangu 09210- 580, Santo André, SP, Brazil usually the largest obstacle to management projects and 3 [email protected] conservation of anuran fauna (Greene, 2005; Silvano * Corresponding Author: [email protected] and Segalla, 2005; Verdade et al., 2012). The aim of 348 Helena Nery Alves-Pinto et al. this study was to identify sexual dimorphism from considering an imaginary mid-line between head and morphometric characteristics in the species Pipa pipa urostile; tibia length (TIL); eye diameter (ED); hand and geographical variation on morphometric variables length (HL); fingers 1 to 4 length (F1 – 4 L); foot length between populations. We also provide notes on its diet (FOL) (Verdade and Rodrigues 2007); mouth (Mt); and gonad development. forearm length (FL) (Márquez-Garcia et al. 2009); forearm diameter (FD) from dorsal to the ventral side Materials and Methods in an imaginary mid-line between elbow and hand; toes 1 to 5 length (T1 – 5 L) from base to the tip of each toe; We studied 61 individuals of Pipa pipa from three interdigital membranes 1 to 4 length (IM1 – 4 L) from different sources: individuals (n=32) collected in the base to the tip considering an imaginary mid-line January 2007 near the left bank of Abacaxis River, São between toes; and snout length (SL) from the tip of the Sebastião village, Borba municipality, Amazonas state, mid-line of the jaw to the tip of the snout. We measured Brazil (4º18’32” S, 58º 38’11” W) (MZUSP 142744– variables on the left side of all individuals, with a digital 775); individuals (n=4) collected in April 2010 at the calliper. right bank of Madeira river, Abunã village, Porto Velho To test sexual dimorphism we used variables’ residuals municipality, Rondônia state, Brazil (9º37’45” S, 65º from a regression analysis in order to avoid biases in 27’19” W) (MZUSP 143285–286, 144395–396); and morphometric comparisons. Levene’s and Shapiro Wilk individuals (n=25) from different localities, deposited at tests were performed to test normality and residual’s the Museu de Zoologia da Universidade de São Paulo variance. A One-way ANOVA was made with normal (MZUSP) collection (MZUSP 1006, 2036, 15903, and homogeneous data, whereas a Mann-Whitney U 16066, 16068, 23062, 23119, 23215, 23374–76, 23795, test was performed with the nonparametric residuals. 37519, 54799–800, 58534–35, 80450, 80454–55, We studied morphological geographic variation using 80474–78). all three samples: from São Sebastião, Abunã and The sample from São Sebastião village (Abacaxis previously specimen deposited at MZUSP, totalizing River) was assembled from a pond approximately six 61 individuals (one young excluded from the original meters long by six meters wide and 30 centimetres deep, sample) of Pipa pipa from 12 different localities. We in an area susceptible to seasonal flooding. The water performed Principal Component Analysis to explore was muddy, and leaf litter and mud covered the bottom. any group formation based on morphometric variation. The stream where individuals from Abunã village All variables listed earlier in this study were included were found was approximately one meter deep and in the analysis. Statistics was performed in SPSS 20.0 three meters wide, and had similar aspects as the São for Windows. Sebastião pond. All individuals were euthanized using a lethal dose of anaesthetic , preserved in 10% formalin Gonad development and stored in 70% ethanol. Gonadal studies were restricted to the sample of São Morphometric studies Sebastião, since we could not examine internal anatomy from the other populations. Testis length, diameter and We tested sexual dimorphism in 32 individuals from volume were measured and compared to male’s weight the São Sebastião sample, in which sex was determined and size using correlation analysis. Diameter, weight by gonad examination. The other samples were not and number of developed ova were measured. included in this analysis since gonad examination was not possible. Sexual maturity was determined based on Diet the snout-vent length (SVL) and gonad development. According to Schuette and Ehrl (1987), only individuals Dietary studies were also restricted to the sample of larger than 60 mm are considered adults. Although São Sebastião, since we could not remove stomach sexual maturity may vary from one population to contents from the other populations. The approximated the other, we still considered this size limit since all time between collection of and fixation was of individuals were larger than 60 mm with the exception one day. All individuals had their gastric and intestinal of one, a 36 mm juvenile. contents analysed under a stereoscope microscope. Prey Thirteen measurements were obtained and analysed length and diameter were measured, and prey category in this study: snout-vent length (SVL); body height identified to the level of Order. Degraded items were (BOH) taken from dorsal to the ventral side of the body not considered because they could not be identified. Morphometric variation of Pipa pipa with notes on diet and gonad development 349

Figure 1. Geographic distribution of Pipa pipa based on literature data (Vaz-Silva and Andrade 2009) and on our data.

The frequency of appearance of each prey, as well as (p= 0.650, F=0.210), but forearm diameter (FD) (p=0.00 the occurrence of stomachs that contained each prey F=26.833), finger 2 length (p=0.00; F=12.522) and was determined. Prey volume was approximated to snout length (SL) (p=0.03, F=10.614) were significantly the ellipsoid volume, calculated based on the formula different between males and females. SL represented of a prolate spheroid V = 4/3π * L/2 * (W/2)2, where between 3.4 to 4.8% of SVL in females, while in males V=volume, L=length, and W=width. it ranged from 3.8 to 5.4% of SVL. To investigate the geographic variation in Results morphometric variables of Pipa pipa, data were taken from 12 localities (Figure 1), nine of which were not Morphometric studies registered by Vaz-Silva and Andrade (2009), who We found 17 males and 15 females in the São summarized the distribution of the species. Differences Sebastião sample. The One-way ANOVA showed that between groups did not clearly split populations. males and females do not differ on snout-vent length According to the Principal Component Analysis, the 350 Helena Nery Alves-Pinto et al.

Figure 2. Plot of scores of the first and second components for the Principal Component Analysis, showing the geographic variation of different populations of Pipa pipa, based on morphological characters.

first component explained 47.98% of the variation, and abundant than other items and had the highest frequency the second explained 26.93%. SVL (0.846), mouth, (Mt) (67%), followed by Diptera larvae (7.3%) and (0.838) and foot length (FOL) (0.834) were the variables Heteroptera (6.3%). Heteroptera showed the highest that most contributed to the separation of groups in the occurrence in stomachs. Even though decapods were not score 1, whereas finger 2 length (F2L), finger 1 length abundant (3.1%), they represented the higher volume (F1L) and finger 4 length (F4L) (0.721; 0.715; 0.712) (61.9 mm3), whereas Cyclopoida represented only 8.5 contributed the most in score 2 (Figure 2). mm3 (Table 2). Skin fragments were found in three stomachs, and their appearance could either indicate Gonad development these have been from an anuran of the genus Rhinella Fitzinger 1926 or from its own individual of Pipa pipa, Testis volume increased positively and proportionally since many are known to eat their own skin with weight (r2 = 0.909, p=0.035) and SVL (r2 = 0.807, (dermatophagy) (Weldon et al. 1993). p=0.001). The two mature females (with developed ova, ready to reproduce; MZUSP 142751, 142768) Discussion had large, yellow ova, in which the vegetative and the poles had the same colour (Figure 3, Table Morphometric studies 1). Inside the ovary, smaller ova were found, and the Forearm diameter and snout length were the only oviduct occupied a great portion of the visceral cavity. sexually dimorphic characteristics in Pipa pipa, both In immature females the ovary was light yellow and larger in males. Forearm diameter was previously long, parallel to the kidneys. reported in the literature as sexual dimorphic in Pipa pipa (Zippel, 2006), but the longer snout is a novelty that Diet may help to externally distinguish sex of individuals. Only 15 out of the 32 digestive tubes examined According to Zippel (2006), literature shows that SVL contained identifiable items (Table 2). Diet contained is larger in females, whereas we found no differences mainly arthropods, of which Cyclopoida was more between sexes. These contradictory results may be due Morphometric variation of Pipa pipa with notes on diet and gonad development 351

to the age of the individuals in the samples studied: the size of the ovule found in females from São Sebastião were smaller than the fully developed ones described in literature (Rabb and Snedigar, 1960), and therefore probably under development (Monnet and Cherry, 2002). In some anurans males are significantly larger than females, especially when males are territorial and fight each other (Shine, 1979; Woolbright, 1983). In Pipa pipa males are extremely territorial and fight frequently. Rabb and Rabb (1963) described the vocalization and physical contact during a fight, in which mainly the head and forearm were used. This might explain the larger forearm diameter in males from São Sebastião. Males with hypertrophied forearms are common among anurans (Duellman and Trueb, 1994), and Liao et al. (2012) showed that forelimbs tend to be stronger in amplectant male specimens of the species Bufo andrewsi, helping to keep the grip on the female during Figure 3. Mature female of Pipa pipa with developed ova. amplexus. SVL = 110 mm. Some dimorphic structures may be used in competition for mates between males as a strategy during contests, pushing or holding the contender. However, in some cases the structures may indicate male’s size, strength Table 1. Data from the mature females from São Sebastião and fitness without being effectively used in combat (Willemart and Gnaspini, 2003). Snout length is an sample.Table 1. example of a structure that is not directly used in combat, Field Average SVL Weight Number of Weight of all eggs Number diameter of eggs but that could indicate size superiority. (mm) (g) eggs (g) (MZUSP) (mm) The morphological variation among samples studied 142751 110 85 74 3.856 4.26 may be suggestive of geographic variation. Nevertheless, 142768 107 110 101 4.614 4.02 considering our limited geographic covering, this could only be confirmed with additional sampling comprising a larger number of individuals per locality.

Gonad development Table 2. Description of preys found as part of the diet in Pipa Rabb and Snedigar (1960) found ova with an average pipa. (N) number of individuals; (%n) frequency; (Oc) number diameter of 6 mm in females that spawned between 96 of stomachs with that prey; (%Oc) number of stomachs with and 273 eggs. Even though the number of ova recorded thatTable prey 2. per total of stomachs; (∑ Vol.) sum of the volumes. in the literature is similar to the number we found, their volume Class Order N %n Oc %Oc sizes suggest that females from this study were not fully (mm3) ready for reproduction. Smaller ova as well as the low Amphibia Anura (skin) 3 3.1 3 15.0 5.4 Insecta Heteroptera 6 6.3 3 15.0 3.4 number of reproductive females in our sample may be Diptera 3 3.1 3 15.0 explained either by the seasonality (early rainy season, Homoptera 2 2.1 2 10.0 6.2 when this species reproduce), or due to young age of the Hymenoptera 4 4.2 4 20.0 population sample. Coleoptera 1 1.1 1 5.0 Orthoptera 1 1.1 1 5.0 Diptera larvae Diet 7 7.3 2 10.0 6.3 (chironomidae) Aracnida Acari 1 1.1 1 5.0 0.0 According to the results, Pipa pipa has an opportunist Crustacea Decapoda 3 3.1 3 15.0 61.9 diet related mainly to the aquatic environment. Copepoda Cyclopoida 64 67.0 5 25.0 8.5 Literature recorded animals captured in nature to feed Total = 95 on fish - Erytrinus erytrinus (Bloch and Schneider 352 Helena Nery Alves-Pinto et al.

1801) (Duelman, 1978) and a frog genus Leptodactylus borealis in Taita Hills: effects of habitat and predator size. Fitzinger 1826 (Deckert, 1917). In captivity it was African Journal of Ecology 48: 299-303. recorded to feed on different types of items, from Buchacher, C.O. (1993): Field studies of the small Surinam toad, Pipa arrabali, near Manaus, Brazil. Amphibia-Reptilia 14: 59- bovine meet to copepods (Rabb and Snedigar, 1960; 69. Rabb, 1969; Palmer, 1994). Canedo, C., Garcia, J.P., Fernandes, R., Pombal, Jr.J.P. (2006): The little amount of preys found in the sample studied Diet of Pipa carvalhoi (Amphibia, Pipidae) is not influenced by in this work may be a consequence both of the gap time female parental care. Herpetological Review 37: 44-45. between the animals were collected and euthanized, Cannatella, D.C., Trueb, L. (1988): Evolution of pipoid toads: allowing the preys to be digested; or scarcity of prey intergeneric relationships of the aquatic frog family Pipidae items due to the season, since the water level of the (Anura). Zoological Journal of the Linnean Society 94: 1-38. ponds may influence the frequency and availability Carreno, C.A., Nishikawa, K.C. (2010): Aquatic feeding in pipid : the use of suction for prey capture. The Journal of of prey in the habitat, as well as competition among Experimental Biology 213: 2001-2008. individuals (Deckert, 1917; Duelman, 1978; Palmer, Cuello, M.E., Bello, M.T., Kunn, M., Úbeda, C.A. (2006): 1994). Nevertheless, composition of Pipa pipa`s diet Feeding habits and their implications for the conservation of was similar to other species in the same genus: P. the endangered semiaquatic frog Atelognathus patagonicus aspera, P. arrabali, and P. carvalhoi were all recorded (Anura, Neobatrachia) in a Northwestern Patagonian pound. to have ingested arthropods, such as nymphs and larvae Phyllomedusa 5: 67-76. of insects, and aquatic invertebrates, as well as flying Deckert, R.F. (1917): Pipa americana rediscovered on Trindad. and terrestrial prey (Trueb and Massemin 2000, Canedo Copeia 1917: 49-50. Duellman W.E. (1978): The biology of an equatorial herpetofauna et al., 2006; Garda et al. 2006). The latter may indicate in amazon Equador. Miscellaneous Publication Museum of either that these fall into the pond or that these species Natural History University of Kansas 65: 1-352. go forage in land (Buchacher, 1993). Duellman, W.E., Trueb, L. (1994): Biology of Amphibians. Baltmore: Johns Hopkins University Press. Conclusions Dunn, E. (1948): American frogs of the family Pipidae. American museum Novitates 1384: 1-13. This study presented some new information about Frost, D.R., Grant, T., Faivovich, J., Bain, R.H., Haas, A., Haddad, the species Pipa pipa, such as new sexually dimorphic C.F.B., de Sá, R.O., Channing, A., Wilkinson, M., Donnellan, morphometric structures, as well as its occurrence in new S.C., Raxworthy, C.J., Campbell, J.A., Blotto, B.L., Moler, geographic locations. Furthermore, it added preliminary P., Drewes, R.C., Nussbaum, R.A., Lynch, J.D., Green, D.M., information about geographic morphometric variation Wheeler, W.C. (2006): The tree of life. Bulletin of among populations, suggesting further studies with the American Museum of Natural History 297: 1-370. Frost, D.R. (2011): Amphibian Species of the World: an online larger samples should be informative. Finally, although reference. Version 5.5 (31 January, 2011). Electronic Database seasonality may have interfered in our sample, more available at http://research.amnh.org/herpetology/amphibia/ information was provided regarding Pipa pipa`s diet index.php. American Museum of Natural History, New York, and gonad characteristics. USA. Last accessed on 18 March 2013. Garda, A.A., Bivati, G.M., Costa, G.C. (2006): Sexual dimorphism, Acknowledgements. We thank J. M. Guellere, J. Cassimiro, M. female fertility, and diet of Pipa arrabali (Anura, Pipidae) in Teixeira Jr, S. M. Sousa, P. Nunes and M. Sena for help in field; Serra do Cachimbo, Pará, Brazil. South American Journal of C. Mello and H. Zaher for the access to MZUSP collection. We Herpetology 1: 20-24. are very thankful for S. Bueno, D. Caetano and P. Fuentes for Greene, H.W. (2005): Organisms in nature as a central focus for the identification of diet items. Teresa Ávila and an anonymous biology. Trends in Ecology and Evolution 20: 23-27. reviewer for revising the article. FAPESP and CNPq provided Irisarri, I., Vences, M., Mauro, D.S., Glaw, F., Zardoya, R. (2011): funding for this project. Authorization for collection was given Reversal to air-driven sound production revealed by a molecular by IBAMA to MTR (Proc. 10126-1). phylogeny of tongueless frogs, family Pipidae. Evolutionary Biology 11:114. References Liao, W.B., Wu, Q.G., Barret, K. (2012): Evolution of sexual dimorphism in the forelimb muscles of Andrew’s toad (Bufo Amaral, A., Rebelo, R. (2012): Diet of invasive clawed frog Xenopus andrewsi) in response to putative sexual selection. Animal laevis at Lage stream (Oeiras, Portugal). The Herpetological Biology 62: 83-93. Journal 22: 187-190. Márquez-García, M., Correa-Solis, M., Sallaberry, M. and Blaustein, A.R., Kiesecker J.M. (2002): Complexity in conservation: M.A. Méndez. Effects of pond drying on morphological and lessons from the global decline of amphibian populations. life-history traits in the anuran Rhinella spinulosa (Anura: Ecology Letters 5: 597-608. Bufonidae). Evolutionary Ecology Research 11: 803-815. Beryl, A.B., Measey, G.J. (2010): Diet composition of Xenupus Measey, G.J. (1998): Diet of feral Xenopus laevis (Daudin) in Morphometric variation of Pipa pipa with notes on diet and gonad development 353

South Wales, UK. Journal of Zoology 246: 287-298. Monnet, J.M. and Cherry, M.I. (2002): Sexual size dimorphism in anurans. Proceedings of the Royal Society of London series B- Biological Sciences 269: 2301-2307. Palmer, C.B. (1994): Adventures with Pipidae: the Surinam toad. Captive Breeding 2: 22-25. Rabb, G.B. (1961): The Surinam toad. Natural History 70: 40-45. Rabb, G.B. (1969): Frogs and pipid frogs. Brookfield Bandarlog 37: 3. Rabb,G.B., Rabb, M.S. (1960): On the mating and egg-laying behaviour or the Surinam toad, Pipa pipa. Copeia 1960: 271- 276. Rabb, G.B., Rabb, M.S. (1963): Additional observations on breeding behaviour of the Surinam toad, Pipa pipa. Copeia 1963: 636-642. Rabb, G.B., Snedigar, R. (1960): Observations on breeding and development of the surinam toad, Pipa pipa. Copeia 1960: 40- 44. Schuette, F., Ehrl, A. (1987): Zur Haltung und Zucht der grossen sudamerikanishen Wabenkrote Pipa pipa. Salamandra 23: 256- 268. Shine, R. (1979): Sexual selection and sexual dimorphism in the amphibian. Copeia 1979: 297-306. Silvano, D.L., Segalla, M.G. (2005): Conservação de anfíbios no Brasil. Megadiversidade 1: 79-86. Trueb, L., Cannatella, D.C. (1986): Systematics, morphology and phylogeny of the genus Pipa (Anura, Pipidae). Herpetologica 42: 412-449. Trueb, L., Massemin, D. (2000): The osteology and relationships of Pipa aspera (Amphibia: Anura: Pipidae), with notes on its natural history in French Guiana. Amphibia-Reptilia 22: 33-54. Vaz-Silva, W., Andrade, T.A. (2009): Amphibia, Anura, Pipidae, Pipa pipa: distribution extension, new state record and geographic distribution map. Check List 5: 507-509. Verdade, V.K., Rodrigues, M.T. (2007). Taxonomic Review of Allobates (Anura, Aromobatidae) from the Atlantic Forest, Brazil. Journal of Herpetology 41 (4): 566-580. Verdade, V.K., Carnaval, P.H., Schiesari, L., Toledo, L.F., Mott, T., Andrade, G., Eterovic, P., Menin, M., Pimenta, B.V.S., Lisboa, C.S., Paula, C.D., Silvano, D.L. (2012): A leap further: the Brazilian Amphibian Conservation Action Plan. Alytes 29: 28-43. Weldon, P.J., Demeter, B.J., and Rosscoe, R. (1993). A survey of shed skin-eating (Dermatophagy) in Amphibians and Reptiles. Journal of Herpetology 27 (2): 219-228. Willemart, R.H., Gnaspini, P. (2003): Comparative density of hair sensilla on the legs of cavernicolous and epigean harvestmen (Arachnida, Opiliones). Zoologischer Anzeiger 242: 353-365. Woolbright, L.L. (1983): Sexual selection and size dimorphism in anuran amphibian. American Naturalist 121:110-119. Zippel, K.C. (2006): Further observations of oviposition in the Surinam toad (Pipa pipa), with comments on biology, misconceptions, and husbandry. Herpetological Review 37: 60- 68.

Accepted by Mirco Solé