Testing for adaptive radiation: a new approach applied to Madagascar frogs

Adaptive radiation is a key topic at the intersection of ecology and
evolutionary biology. Yet the definition and identification of adaptive
radiation both remain contentious. Here, we introduce a new approach for
identifying adaptive radiations which combines key aspects of two widely
used definitions. Our approach compares evolutionary rates in morphology,
performance, and diversification between the candidate radiation and other
clades. We then apply this approach to a putative adaptive radiation of
frogs from Madagascar (Mantellidae). We collect new data on morphology and
performance from mantellid frogs and compare rates of diversification and
multivariate evolution of size, shape, and performance between mantellids
and other frogs. We find that mantellids potentially pass our test for
accelerated rates of evolution for shape, but not for size, performance,
or diversification. Our results demonstrate that clades can have
accelerated phenotypic evolution without rapid diversification (dubbed
“adaptive non-radiation”). We also highlight general issues in testing for
adaptive radiation, including taxon sampling and the problem of including
another adaptive radiation in the comparison clades. Finally, we suggest
that similar tests should be conducted on other putative adaptive
radiations on Madagascar, comparing their evolutionary rates to those of
related clades outside Madagascar. Based on our results, we speculate that
older Madagascar clades may show evolutionary patterns more similar to
those on a continent than an island.