Macroecology and extinction risk correlates of frogs

Abstract

Aim Our aim was to test whether extinction risk in frogs could be predicted from their body size, fecundity or geographic range size. Because small geographic range size is a correlate of extinction risk in many taxa, we also tested hypotheses about correlates of range size in frogs.

Methods Using a large comparative dataset (n = 527 species) compiled from the literature we performed bivariate and multiple regressions through the origin of independent contrasts, to test proposed macroecological patterns and correlates of extinction risk in frogs. We also created minimum adequate models to predict snout-vent length, clutch size, geographic range size and IUCN Red List status in frogs. Parallel non-phylogenetic analyses were also conducted. We verified the results of the phylogenetic analyses using gridded data accounting for spatial autocorrelation.

Results The most threatened frogs tend to have small geographic ranges although the relationship between range and extinction risk is not linear. In addition, tropical frogs with small clutches have the smallest ranges. Clutch size was strongly positively correlated with geographic range size (r2 = 0.22) and body size (r2 = 0.28).

Main conclusions Our results suggest that body size and fecundity only affect extinction risk indirectly through their effect on geographic range size. Thus, although large frogs with small clutches tend to be endangered, there is no comparative evidence that this relationship is direct. If correct, this inference has consequences for conservation strategy: it would be inefficient to allocate conservation resources on the basis of low fecundity or large body size; instead it would be better to protect areas which contain many, small geographic range frog species.