The degree of frugivory of birds as estimated from gastric and fecal samples

Abstract

The degree of frugivory (DF) has long been used to characterize the relative importance of fruits in the diet of a bird and, more recently, as a functional trait related to the role of birds in seed dispersal networks. Although quantitative estimations of DF are desirable, general, categorical classifications or coarse estimations of diet composition based in qualitative information are often used. Data on stomach, fecal, and regurgitation contents scattered in the literature or easily obtained in the field could be used to provide a quantitative, potentially more reliable assessment of DF. We compiled such data from the literature and our own fieldwork to obtain 12,576 samples from 985 Neotropical bird species, of which 489 species (49.6%) in 61 families had at least one fruit-containing sample. From this dataset we analyzed 250 species for which we got at least 10 samples to find that gastric (i.e., stomach plus regurgitation) and fecal samples provided similar estimates of DF despite potential differences in the degree of food digestion. The DFs we obtained were higher than those presented in the most frequently used source of quantitative DF estimates in the literature (the Elton Traits database). We further explored, with a few study cases, the utility of stomach, fecal and regurgitation samples to evaluate intraspecific geographic, sexual, and ontogenetic variations in DF, topics rarely investigated so far. We found that the range of intraspecific geographic variation in DF increased with increasing DF, and that sexual and ontogenetic differences may occur. We argue that stomach, fecal, and regurgitation sample data abundantly available in the literature or easily obtained from mist-netted birds may be used to produce quantitative assessments of DF likely more reliable than the estimates used so far and useful for a plethora of ecological studies.