Diet and trophic niche overlap of Boana bischoffi and Boana marginata ( Anura : Hylidae ) in southern Brazil

The species Boana bischoffi and Boana marginata are endemic to the Atlantic Forest, where they often occur in sympatry. There is a large gap in the knowledge of natural history of both species. In this study, we aimed to describe and compare the diet composition of B. bischoffi and B. marginata in the southern domain of the Brazilian Atlantic Forest. We analyzed the gastrointestinal contents of 43 individuals of B. bischoffi and 30 individuals of B. marginata. Both showed a high trophic niche overlap (0.90 Ojk). The most important prey categories for both species belonged to the orders Araneae and Coleoptera. The species niche breadth (Bsta) varied from 0.35 to 0.42, suggesting a generalist feeding behavior for both species. Our data provide unprecedented information on these species’ food composition, contributing to a better knowledge of the natural history of neotropical anurans.


Introduction
Information on species trophic ecology is fundamental to describe their ecological niches (Sih 2001), which, in turn, are necessary to understand the functioning of communities and ecosystems (Solé & Rödder 2010).Several studies have been carried out to investigate the diet of neotropical anurans (López et al. 2009, Sabagh et al. 2012, Barbosa et al. 2014, Kittel & Solé 2015, Oliveira et al. 2017), however, few evaluate the trophic relationship of resource-sharing species (Sabagh et al. 2010, Oliveira et al. 2015, Moser et al. 2017).Studies on the trophic ecology of coexisting species can provide important information about the use of food resources, thus helping to understand ecological interactions (Begon et al. 2007).
The Hylidae family has 714 known species (Frost 2018) and, in the extreme south of Brazil, they feed mainly on arthropods of the orders Coleoptera, Araneae and Hemiptera (Kwet 2001).The tree frogs Boana bischoffi (Boulenger, 1887) and Boana marginata (Boulenger, 1887) are endemic to the Atlantic Forest biome (Marcelino et al. 2009) and occur in sympatry in the southern Brazilian plateau.Both species have a similar biology but B. bischoffi occurs both in forest and open areas, while B. marginata occurs exclusively in closed forest areas (Kwet et al. 2010).
attached to the bottom of the Petri dish, where we evenly spread each item, maintaining a regular height of 1 mm (Hellawell & Abel 1971).In order to calculate each item volume (V), the area value (mm²) was multiplied by its height (1 mm) (Oliveira et al., 2015).

Data Analysis
To measure the importance of each prey category, we used the Index of Relative Importance (IRI): IRI = (% N +% P) % FO, where %N is the relative abundance of each prey in the diet; %P is the mass percentage of each prey in the diet; and %FO is the relative frequency of occurrence of prey (Pinkas et al. 1971, Krebs 1999).The higher the value of IRI, the greater is the importance of each prey in the diet.To analyze the dimension of the trophic niches breadth, we calculated the Levin's Niche Breadth Index (B) (Krebs 1999).This index is defined by: B = 1/Σpi2, where p is the individual proportion of a given resource i (taxon) found in the diet.To facilitate comparisons between species, the Standardized Index of Levins (Bsta) was calculated using the following equation: Bsta = (B-1) / (n-1), where n represents the number of resources recorded.Values close to zero are attributed to a specialist diet, while values close to one are attributed to a generalist diet.
To analyze if the trophic niche overlaps between the species, with regards to the degree of similarity between their diets, we used the Trophic Niche Overlap Index of Pianka (Ojk) (Pianka, 1974), defined by the following equation: , where Ojk is the niche overlap index between the species j and k; pij is equivalent to the proportion of the resource type i relative to the total of resources used by the species j; pik is the proportion of resource i relative to the total of resources used by the species k; and n is the total number of resource categories used by the species j and k.The index ranges from 0 to 1, when there is no overlap or a complete overlap between the species diets, respectively (Krebs, 1999).

Results
We analyzed the gastrointestinal contents from 43 individuals of B. bischoffi and 30 individuals of B. marginata.Of these, only two individuals (both B. marginata) had their digestive tracts empty.We identified a total of 149 food items, distributed in 18 prey categories.We identified 15 prey categories for B. bischoffi and 12 for B. marginata.According to the Index of Relative Importance (IRI), the two most important prey for both B. bischoffi and B. marginata were Araneae (IRI% = 45.2 and 29.1, respectively) and Coleoptera (IRI% = 25, 7 and 58.6, respectively) (Table 1).The "others" group represented 15.9% of the total volume in B. bischoffi and 38.9% in B. marginata.In both species, we found plant material, representing 2.30% of the total volume of B. bischoffi and 5.10% of B. marginata (Table 1).Only three prey categories were consumed exclusively by B. marginata: Odonata, Isopoda and Formicidae.The number of exclusive prey was higher for B. bischoffi: Lepidoptera, Lepidoptera (larva), Dermaptera, Diptera (larva), Pseudoescorpionida and Diplopoda.However, the two species had a high overlap in their diet, according to the Pianka Niche Overlap Index (0.90 Ojk).In relation to niche breadth, the Levins' Standardized Trophic Niche Index (Bsta) was higher for B. bischoffi (0.42) than for B. marginata (0.35) (Table 1).
Morphologically and phylogenetically close species tend to use resources in a similar way (Ricklefs & Trevis 1980) and understanding how this occurs is essential to answer questions about community structure (Lawor 1980).Effective and widely used parameters to understand the species trophic relation are the trophic niche breadth and niche overlap (Pianka 1974, Sabagh & Carvalho-E-Silva. 2008, Sabagh et al. 2010).In this study, we describe and compare the food composition of B. bischoffi and B. marginata, as well as their trophic niches, to provide information about the natural history of these hylids.

Study site
The study was carried out at the southern boundary of the Atlantic Forest, in a region of Mixed Ombrophilous Forest (or Araucaria Forest).Samplings were carried out in a well-preserved portion of Araucaria Forest, inside a Federal Conservation Unit, the National Forest of São Francisco de Paula (29°25'22.4''S;50°23'11.2''W,907 a.n.m.), municipality of São Francisco de Paula, state of Rio Grande do Sul, Brazil.The reproductive site sampled was formed by a lake connected to a small stream, where both B. bischoffi and B. marginata were found (Figure 1).The area has an annual precipitation of 2162 mm distributed evenly throughout the year, subtemperate climate and with an annual average temperature of 14.5 °C (Backes 1999, Maluf 2000).

Data Collection
Captures were made through active search (Crump & Scott Jr 1994) at night from October to November 2015.The individuals found were manually captured and immediately packed in a refrigerated styrofoam box to reduce their digestion's physiological activities (Oliveira et al. 2015).The collections were carried out with the competent Federal Organ authorization, SISBIO (authorization # -45861-1).At the laboratory, the animals were euthanized with topical anesthetic (xylocaine), fixed in 10% formaldehyde and preserved in 70% alcohol.This procedure was in accordance with the institutional license obtained for experimentation with animals (# PPCEUA07.2016).Subsequently, each individual was dissected for the removal of gastrointestinal contents (stomach and intestine considered as a single sample).This material was kept in 70% alcohol until the screening process with a stereomicroscope.In addition to food content, we also collect muscle tissue from individuals for stable isotope analysis, which was part of a complementary study developed by our staff.All collected individuals are housed in the scientific collection of the Laboratory of Ecology of Terrestrial Vertebrates (Boana bischoffi: CHLEVT 1248CHLEVT -1275CHLEVT , 1324CHLEVT , 1325CHLEVT , 1327CHLEVT , 1328CHLEVT , 1342CHLEVT -1352;;Boana marginata: CHLEVT 1364-1382, 1384-1392, 1703).Extracted contents were identified until the lowest possible taxonomic level, which was limited by prey digestion level.In general, prey presented a high fragmentation degree and, for this reason, the maximum taxonomic resolution reached mostly the level of order.For each prey category, the number, volume (detailed below) and frequency of occurrence were calculated.Some fragments whose identification was not possible were grouped in the category "others" and only their volume was quantified.Volume was calculated by using the area (mm²) occupied by each item with a graph paper support

Discussion
Boana bischoffi and B. marginata exhibited similar composition in their diets.In addition, the most important prey were the same for both species.This similarity in their diets was expected due to phylogenetic proximity between them (Faivovich et al., 2004).Furthermore, it is likely that both species were exposed to the same prey supply since they were collected in the same study site, although in possible different microhabitats.Unfortunately, we do not have data on the prey availability in the habitat, and therefore our results should be interpreted with caution.The taxonomic level of Order could be a limitation for our explanations (Tavares-Cromar & Williams 1996, Bundy et al. 2011, Pombo et al. 2013).Nevertheless, some basic aspects of food-web, as the mean chain length and linkage complexity vary little within the degree of taxonomic resolution (Thompson & Townsend 2000), and food webs are best explained by coarser taxonomic ranks, from kingdom to class (Eklöf et al. 2012).
The orders Araneae and Coleoptera are considered very abundant arthropods in the environment (Baretta 2007) and were also recorded in abundance in the study area (Oliveira M. pers. comm.).The importance of Araneae and Coleoptera has also been recorded for several other neotropical anurans (e.g., Maneyro et al. 2004, Miranda et al. 2006, Duré et al. 2009, Rosa et al. 2011, Oliveira et al. 2015, Moser et al. 2017).In addition, spiders and beetles were also the most important prey for other species of the same genus, such as Boana leptolineata, in a region that is similar and close to the studied environment (Barbosa et al. 2014).Coleopterans are a group of highly diverse organisms and very abundant in the environment, especially during the warmer months (Teixeira et al. 2009), making them prey of easy consumption.

Figure 1 .
Figure 1.Anurans and sampled environments in the Araucaria Forest at the National Forest of São Francisco de Paula, municipality of São Francisco de Paula, state of Rio Grande do Sul, Brazil.A, Boana bischoffi, CHLEVT 1250, adult male, SVL 40.0 mm; B, Boana marginata, CHLEVT 1379, adult male, SVL 50.0 mm; C and D lentic and lotic environments, respectively.