A few years ago I published the results of a phylogenetic analysis of New World Exoristinae based on molecular data from two genes, 28S rDNA and Elongation Factor 1-alpha (Stireman 2002). In that study I employed parsimony, neighbor joining, and maximum likelihood inference methods to generate phylogenetic reconstructions, and explored a variety of weighting schemes and combinations of the sequence data (i.e. each gene separately and both together). The results of these analyses generally supported recent taxonomic hypotheses (e.g., Herting 1984; Wood 1987; O’Hara and Wood 2004). For example, Tachinidae and Exoristinae were reconstructed as monophyletic in most analyses, as were the Exoristini, Winthemiini, and Blondeliini. However, there were also some ambiguous and unexpected results. First, representative taxa of Tachininae and Phasiinae (used as outgroups) failed to support monophyly for either of these subfamilies. Also, species of the genera Masiphya (Masiphyini), Ceracia (Tachininae), and Phyllophilopsis (Blondeliini) tended to form a clade that varied widely in position between reconstructions. Perhaps most interesting, all reconstructions indicated a paraphyletic or polyphyletic Goniini. Finally, and most disturbing, was the fact that my representative of Drino (D. incompta) was often reconstructed near the base of Exoristinae joining taxa from other subfamilies (at least in analyses of EF1 alpha) even though all morphological considerations would place this taxon with other “Eryciini”.
Stireman, J. O.
(2005). Phylogenetic Reconstruction of Exoristinae Using Molecular Data: a Bayesian re-analysis. The Tachinid Times (18), 4-6.