Molecular phylogeny of bark and ambrosia beetles (Curculionidae: Scolytinae) based on 18 molecular markers

Abstract

AbstractThe phylogeny of the large weevil subfamily Scolytinae has been difficult to resolve based on a limited number of genetic markers. With more than 6000 nominal species in the subfamily, the general lack of resolution at deeper nodes indicates that large sequence volumes are needed to solve this problem. We have therefore assembled a large molecular dataset consisting of more than 10 kb of nucleotides from 18 gene fragments, for 182 species. Nucleotide and amino acid translated data were analysed using Bayesian and parsimony‐based approaches, which gave largely congruent results. Compared with previous analyses, we obtained greater resolution for some of the deeper nodes, and detected many unexpected relationships that were strongly supported by our data. The tribe Scolytini was recovered as the earliest divergent lineage in Scolytinae, sometimes placed together with the hexacoline genus Microborus. Among the 26 currently recognized tribes, 15 were monophyletic, whereas the remaining tribes were largely paraphyletic. The majority of species in the tribe Hypoborini were recovered as the sister lineage to a large group containing the species‐rich tribe Dryocoetini, which includes the recently radiated ambrosia beetles in Xyleborini, and Ipini, which includes another recent group of ambrosia beetles in Premnobiina. Cryphalini, Hylesinini and Hylurgini were strikingly polyphyletic tribes, each consisting of several independent lineages. Subgroups were to a large degree defined by geographical affinities, showing a clear distinction between the northern and southern hemispheres. The affiliation of the inbreeding genus Hypothenemus was revealed with strong support as the sister group to the Malagasy and East African species of the genus Cosmoderes. Cryptocarenus was previously assumed to be the sister lineage of Hypothenemus, but was here found to be part of Corthylini, near Araptus. These and many other findings document the need for a thorough revision of the current classification of genera and tribes, including a systematic re‐evaluation of morphological characters.