Diet composition constitutes basic information of the natural history of the species and is necessary to understand one of the requirements for species occurrence and ecological interactions in prey-networks (^{Ceron et al., 2019}). Despite the importance and amount of data acquired on what species eat, much remains to be known on this subject. For instance, when it comes to species with wide geographic range, information on diet from different localities is needed to assess if species have specialized on distinct items at ecologically different regions according to prey abundance and energetic demand (^{Davis et al., 2015}). Anurans plays an important role in trophic networks, and a relatively well studied group concerning latitudinal global patterns of dietary preferences (Ceron et al., 2019). Nevertheless, for most Neotropical anuran species dietary patterns are reported from single populations and through punctual field note, making it difficult to observe patterns on a large geographical scale.

Leptodactylus vastus Lutz, 1930 is a large terrestrial leptodactylid frog (adult specimens' snout-vent length range from 120-180 mm; ^{de Sá et al., 2014}) occurring along open formations from the Cerrado and Caatinga biomes in Brazil and boundaries ofChiquitano forest in Bolivia (^{Heyer, 2005}; ^{Jansen and Schulze, 2012}; de Sá et al., 2014). This species mainly feeds on arthropods and small vertebrates (e.g., frogs and bats; ^{Santana et al., 2012}; ^{Leite Filho et al., 2014}; ^{Caldas et al., 2019}). Herein we report for the first time a predation on the endemic leptodactylid Rupirana cardosoi Heyer, 1999 by a juvenile L. vastus in Bahia state, Brazil, and provide a prey items compilation for L. vastus.

On 4 February 2020 at Chapada Diamantina in the municipality of Mucugê (13°00' S, 41°22' W, 991 m.a.s.l.), Bahia state, Brazil, at 19:25 h, we observed a juvenile preying on R. cardosoi adult male at the margins of a narrow sandy bottom river stream. At the time of observation, we found an individual of L. vastus ingesting its prey first from the back, with only the head outside its mouth (Fig. 1a). In addition to the predated R. cardosoi individual, we also observed several males of that species calling exposed over sandbanks along the margins of this same stream. We manually captured the frog L. vastus (AAGARDA 12852, SVL = 44.17 mm) and placed it in a plastic bag for later procedures in laboratory, when it regurgitated an already dead R. cardosoi (AAGARDA 12851, SVL = 28.91 mm). The prey represented 65.5 % of the predator's SVL (Fig. 1b). We sacrificed the L. vastus with lidocaine 5 %, fixed in 10 % formaldehyde and preserved in 70 % ethanol. Animals are housed at Laboratório de Anfíbios e Répteis at Universidade Federal do Rio Grande do Norte, collected under SISBio permit 19828.

Additionally, we gathered additional data on the prey items of L. vastus in the literature using the combined search terms "Leptodactylus vastus" + "predation", "diet" in the following online sources: Google Scholar (scholar. google.com.br), Scielo (scielo.org), semantic scholar (semanticscholar.org), academic Microsoft (academic. microsoft.com), and Dimensions (app.dimensions.ai). We also included publications that we were aware of and were not detected in these databases, such as publication about frog's community diet.

Our observation is the first record of R. cardosoi, a near threatened species endemic to the Chapada Diamantina (Acuña ^{Juncá and Silvano, 2020}), as a prey item of another species. Due to their small size (Mean = 34.4mm, SD = 1.6, Acuña Juncá and Lugli, 2009), R. cardosoi is probably a prey item of other anuran species and vertebrates of the region, as small frogs are susceptible to being preyed upon by any predator (^{Toledo et al., 2007}). The predation event here reported is probably a consequence of both species are syntopic in the terrestrial strata along stream's margins and due to increased exposure during calling activity of R. cardosoi. However, our observation reports for the first time a juvenile L. vastus preying on another frog.

Figure 1 a) Leptodactylus vastus preying on Rupirana cardosoi backwards at stream margin in the municipality of Mucugê, Bahia state, Brazil. b) Preserved individuals of L. vastus (AAGARDA 12852) and R. cardosoi (AAGARDA 12851). 

We found nine papers addressing diet or predation events of L. vastus reported from the Caatinga (N = 4) and "Mata Atlántica" (the Atlantic Forest; N = 5) biomes (Table 1). Some of the available papers report invertebrate consumption (six categories at the levels of Order and Family), representing an important component in the diet of L. vastus. In all available papers there are reports of vertebrates in the diet of L. vastus, mostly on anurans (seven species) followed by bats (six species) and one species of lizard (Table 1). Batrachophagy by L. vastus seems common (Table 1), being anurans the most important prey item of the species during the dry season in one locality (^{Caldas et al., 2019}). Most anuran preys' reports are leptodactylids, followed by two species of hylids preyed around ponds (Table 1). Leptodactylids are terrestrial species more accessible to L. vastus since they are syntopic (i.e. species present in the same habitat), whereas hylids are often found on a higher vegetal strata, but occasionally forage on the ground and might be opportunistically preyed (^{Haddad et al., 2013}). As for bats, reports show that L. vastus actively preyed on bats trapped into mist nets (see Leite Filho et al., 2014) or positioned themselves near a cave entrance and waited for fallen bats that collided with the narrow entrance (see Gouveia et al., 2009). Our survey reinforces L. vastus as a generalist and opportunistic predator (^{Santana et al., 2012}) of small vertebrates and arthropods (^{Teles et al., 2017}; ^{Caldas et al., 2019}). Although this species is easily detectable and common throughout its wide distribution, we found only one paper (Caldas et al., 2019) reporting L. vastus diet at the same localities for long time span (e.g. one year). These authors observed changes in the importance of some items between wet and dry seasons at different biomes, reporting that L. vastus can even rely on plant material during the dry season (Table 1).

2014) or positioned themselves near a cave entrance and waited for fallen bats that collided with the narrow entrance (see Gouveia et al., 2009). Our survey reinforces L. vastus as a generalist and opportunistic predator (^{Santana et al., 2012}) of small vertebrates and arthropods (^{Teles et al., 2017}; ^{Caldas et al., 2019}). Although this species is easily detectable and common throughout its wide distribution, we found only one paper (Caldas et al., 2019) reporting L. vastus diet at the same localities for long time span (e.g. one year). These authors observed changes in the importance of some items between wet and dry seasons at different biomes, reporting that L. vastus can even rely on plant material during the dry season (Table 1).

All papers reporting batrachophagy of adult L. vastus (Table 1) showed proportionally larger specimens when compared to their preys (e.g. L. vastus, SVL: 114.6 mm, L. latrans, SVL: 75.2 mm; ^{Santana et al., 2012}). Proportional larger body sizes allow predators to easily subdue preys (^{Toledo et al., 2007}). In our observation, specimens do not present a striking difference in body size, but L. vastus was able to almost swallow R. cardosoi. Even though we were not able to measure L. vastus mouth width, this predation event seems to be coherent with previous works highlighting that mouth size is related to selection of prey items in most frogs (^{Coco et al., 2014}). Also, our observation and literature survey reinforce L. vastus as a generalist and opportunistic species.