How much fruit do fruit-eating frogs eat? An investigation on the diet of Xenohyla truncata (Lissamphibia: Anura: Hylidae) H. R. da Silva 1 & M. C. de Britto-Pereira 2 1 Universidade Federal Rural do Rio de Janeiro, Instituto de Biologia – Departamento de Biologia Animal, Serope ´ dica, RJ, Brazil 2 Universidade Federal Fluminense, Instituto de Geocie ˆ ncias – Departamento de Geografia, RJ, Brazil Keywords Hylidae; Xenohyla truncata; diet, frugivory; Restinga; Neoregelia cruenta; bromeliad; Anthurium harrisii; Erythroxylum ovalifolium; Maytenus obtusifolia. Correspondence He ´ lio Ricardo da Silva, Universidade Federal Rural do Rio de Janeiro, Instituto de Biologia – Departamento de Biologia Animal, Caixa Postal 74524, Serope ´ dica, RJ 23851-970, Brazil. Email: helio@ufrrj.br Received 19 April 2005; accepted 3 May 2006 doi:10.1111/j.1469-7998.2006.00192.x Abstract This paper presents the results of a 22-month survey and the examination of the intestinal content of 356 specimens of Xenohyla truncata (Anura: Hylidae) from Restinga de Maric ´a, Rio de Janeiro, Brazil. Our results confirm prior observations that fruits are intrinsic to the diet of these frogs. In addition, these new data increase our understanding of the relationship between frogs and the plants they feed upon. Plant consumption follows availability of fruits in the area, indicating that the diversity of fruits consumed by the frogs does not represent choice, but rather plant phenology and fruit availability. Introduction The hylid frog Xenohyla truncata is unique among frogs with respect to its feeding biology – it includes the fruit of several species of plants in its diet. Silva, Britto-Pereira & Cara- maschi (1989) accidentally discovered this unusual behavior while conducting fieldwork at Restinga de Maric´a, Rio de Janeiro, Brazil. Registers of other plant-eating frogs are scant in the literature and include only two species of Bufo [Alexander, 1964; Zug et al., 1975 (Bufo marinus) and Winston, 1975 (Bufo regularis)] and two species of Rana [Tyler, 1958 (Rana sculenta) and Mondal, 1970 and Das, 1966 (Rana hexadactyla)]. As a food item for frogs, plant material may be more common than reported; nevertheless, as researchers assume it to be ingested accidentally, it is disregarded as important (see the discussion in Hirai & Matsui, 1999). Therefore, that X. truncata demonstrably includes fruit in its diet is by itself a remarkable finding. However, much about the biology and evolution of this behavior remains unknown. This frog that was, for three decades, considered a member of the genus Hyla (Izecksohn, 1959) was placed into a new genus (Izecksohn, 1998) that now includes two species with broadly separate distributions (Caramaschi, 1998). The creation of the genus Xenohyla was based primarily on some oddities of its morphology and behavior; however, a compre- hensive phylogenetic analysis of hylids confirmed the unique- ness of these frogs (Faivovich et al., 2005). In this phylogeny, the genus appears as the sister group to a clade of small-sized hylines now grouped in the genus Dendropsophus. No member of this group is known to eat fruits, although only a few species have been investigated (Parmelee, 1999). Because of the uniqueness of fruit consumption in X. truncata among frogs in general and hylids in particular, understanding the evolutionary origin of this behavior is elusive; even though, based on Faivovich et al. (2005), we could assume that the genus Xenohyla is sister to the genus Dendropsophus, no other frog in this clade is known to have a similar diet (not even the other species of Xenohyla). In addition, hypotheses used to describe evolution of herbivory in lizards (see Espinoza, Wiens & Tracy, 2004; Vitt, 2004) are not directly applicable to frogs. This is mainly because, with a few exceptions, frogs are nocturnal, and cannot use sun heat as the source of extra energy considered necessary for plant digestion among ectotherms because of physiolo- gical limitations. This and other questions related to fruit eating in frogs will remain unanswered until more data is gathered. Herein, we report our findings based on a 3-year survey designed to study the feeding phenology of X. truncata. On the basis of this larger dataset, we can now describe better the variation in the X. truncata diet and relate it to fruit availability. Journal of Zoology 270 (2006) 692–698 c  2006 The Authors. Journal compilation c  2006 The Zoological Society of London 692 Journal of Zoology. Print ISSN 0952-8369