The D2 variable region of 28S ribosomal RNA was sequenced from ethanol specimens or obtained from the literature to provide the first phylogenetic reconstruction of the subfamily Euphorinae (Hymenoptera: Braconidae). Phylogenetic relationships were established by comparing the results using two different methods (distance-based neighbor-joining, NJ; and maximum parsimony, MP) and three different outgroups. The monophyly of the Euphorinae is well supported by all trees generated from molecular data. All phylogenetic reconstructions yielded trees with very similar topologies that only partially resolved the morphologically defined tribes and the relationships within the subfamily. We found no evidence for the monophyletic natures of the tribes Euphorini, Dinocampini, Perilitini, Syntretini, Comsophorini and Centisitini, but we did find some evidence for the tribes Meteorini and Microctonini. The monophyletic nature of the tribe Meteorini was well-supported in all trees. We also found the clade containing the Lecythodella, Microctonus, Orionis and Streblocera to be a monophyletic group, which corresponded to the tribe Microtonini, with Orionis transferred from the tribe Euphorini into Microtonini. Among the genera of Euphorini our results showed strong support for a paraphyletic nature of this group, which can be roughly divided into two clades, one consisting of Aridelus + Wesmaelia, the other of Leiophron + Peristenus, suggesting both of which may be given tribal rank. The placement of the genus Chrysopophorus is largely uncertain. Two clades, Dinocampus + Perilitus and Cosmophorus + Rhopalophorus, were constantly resolved in our analyses, with 42-96 and 97-100 bootstrap value support, respectively, suggesting that both of them form monophyletic groups. For members of the Centistini, Pygostolus may be removed and included in Microctonini or other relative tribe.
