Molecular data resolves relationships within Heteroceridae (Coleoptera: Dryopoidea)

Abstract

Although most authors have recognized five or fewer genera in the Heteroceridae, the most recent revision divided the heterocerids into 19 genera. Under this system most males can be identified to genus level based on genital morphology, but clearly females cannot be identified to genus level by this method. We conducted the first phylogenetic analysis of the family using separate and combined 28S and EF-1α sequence data, with the specific aim of testing generic limits and determining whether former family classifications are natural. We found that Tropicus Pacheco forms a strongly supported sister group to all other Heteroceridae. Augyles s.l. Schiodte (monophyly unresolved) is nested between Tropicus and a large terminal clade of four monophyletic species groups of Heterocerus s.l. Fabricius, which is consistent with trends in morphology and geography. The synonymy of North American Lapsus tristis (Mannerheim) and European Heterocerus fenestratus Thunberg is supported. Tests comparing tree topologies, along with strong evidence that many of the proposed genera are polyphyletic, indicate that the more recent generic circumscriptions are generally unnatural, with the exception of Tropicus, which forms a highly supported clade in all our analyses. We propose that the traditional classification of heterocerids, consisting of the four genera Augyles, Heterocerus, Micilus Mulsant & Ray and Elythomerus Waterhouse, plus Tropicus, is the most appropriate for the family. Molecular dating analyses suggest that vicariance played a major role in diversification, with the major clades of Heteroceridae corresponding to continental distributions and an apparent radiation of Heterocerus s.l. in the early Palaeogene period. Behavioural evidence suggesting a possible role for sexual selection during heterocerid evolution is discussed.