The frog filter: Amphibian introduction bias driven by taxonomy, body size and biogeography

Aim: Invasive species often exhibit a highly non-random suite of traits relative to non-invasive taxa, and these biases reflect strong selection at a series of steps along the invasion pathway. Here we investigate traits that are favoured in the first of these steps: the introduction of species outside their native geographic range. We use the global introduction of amphibians as our case study. Location: Global. Methods: We examined whether taxonomy, body size and attributes of the native geographic range predict the likelihood of an introduction event in amphibians. We then determined whether these characteristics differed between amphibian species that have been introduced intentionally and those that have been introduced unintentionally (i.e. stowaways). Results: Comparisons of introduced and non-introduced amphibians reveal significant biases with respect to taxonomic position (c. 50% of all introduced species originate from only 5% of all families), and characteristics of the native geographic range (most introduced species originate from the Northern Hemisphere, occupy a wide geographic range and are sympatric with high densities of humans). Many of these biases presumably reflect the ease and likelihood of capturing and transporting such animals. Additionally, intentionally introduced species are of larger than average body size, occupy higher elevations in their native ranges, and are more likely to originate from the Northern Hemisphere than unintentionally introduced species. Main conclusions: Introduced amphibian species are not a random subset of the global amphibian fauna with respect to taxonomic affiliation and biogeography, and this restricts our ability to make robust generalizations regarding the ecological determinants of introduction success. Nevertheless, many of our findings are similar to those revealed by previous analyses of vertebrate and invertebrate introductions. Thus, our study suggests that biases in the anthropogenic mechanisms involved in transporting species around the globe are surprisingly consistent across broadly divergent taxa.