Erik M. Pilgrim, Carol D. von Dohlen & James P. Pitts 2008
Molecular phylogenetics of Vespoidea indicate paraphyly of the superfamily and novel relationships of its component families and subfamilieshttp://www3.interscience.wiley.com/jour ... 1&SRETRY=0
The 24 000+ described species of Vespoidea include many well-known stinging wasps, such as paper wasps and hornets (Vespidae), velvet ants (Mutillidae), spider wasps (Pompilidae) and ants (Formicidae). The compelling behaviours of vespoids have been instrumental in developing theories of stepwise evolutionary transitions, which necessarily depend on an understanding of phylogeny, yet, existing morphological phylogenies for Vespoidea conflict. We collected molecular data from four nuclear genes (elongation factor-1α F2 copy, long-wavelength rhodopsin, wingless and the D2–D3 regions of 28S ribosomal RNA (2700 bp in total)) to produce the first molecular phylogeny of Vespoidea. We analysed molecular data alone and in combination with published morphological data from Brothers and Carpenter. Parsimony analyses left many deeper nodes unsupported, but suggested paraphyly of three families. Total-evidence Bayesian inference produced a more resolved tree, in which the monophyly of Vespoidea was nevertheless ambiguous. Bayesian inference of molecular data alone returned a well-resolved consensus with posterior probabilities of over 95% for most nodes. We used this topology as the best estimate of phylogeny at the family and subfamily levels. Notable departures from previous estimates include: (i) paraphyly of Vespoidea resulting from the nesting of Apoidea within a lineage comprising Formicidae, Scoliidae and two subfamilies of Bradynobaenidae; (ii) paraphyly of Bradynobaenidae, Mutillidae and Tiphiidae; (iii) a sister relationship between Rhopalosomatidae and Vespidae; and (iv) Rhopalosomatidae + Vespidae as sister to all other vespoids/apoids. We discuss character evidence in light of the new phylogeny, and propose a new classification of Aculeata that recognizes eight superfamilies: Apoidea, Chrysidoidea, Formicoidea, Pompiloidea, Scolioidea, Tiphioidea, Thynnoidea and Vespoidea
They only used sequences from 4 genes, but still its gonna b useful.
But the amout of samples of the Vespidae, Formicidae, Apoidea and Chrysidoidea are quite small - interesting/new findings of them regard the position of Apoidea within the Vespoidea near Scoliidae and the Chrysididae/Formicidae, so they certainly should have increased the sample size there. They themselves write "Taxon sampling strategy is an important component of phylogenetic analyses, because outcomes are often sensitive to the number and identity of taxa". Chrysidoidea were used as outgroup, but its relatively close related so its maybe prone to errors, therefore a larger samplesize should be used here, or (or better as an addition) a more far outgroup like some Ichneumonidae or such.
Interesting was this note:
"Certain interesting relationships found in the MP (Maximum parsimony) tree, but not supported by the bootstrap, were the following: Chrysidoidea was paraphyletic due to the position of Hedychridium (Chrysididae) as sister to the ants (Formicidae). Vespoidea were rendered paraphyletic by the position of Hedychridium, and by Apoidea, which nested inside Vespoidea as sister to Scoliidae."
The output of that one analysis, the position of the Hedychridium is clearly wrong. There is absolutely no doubt that Hedychridium is part of the Chrysididae and not close to the Formicidae. Seems they solved that problem in another analysis.
The proposed position of Apoidea and their finding for the Mutillidae/Tiphiidae are interesting, but it should be supported in another study with larger sample size and more genes. There is still much to resolve in this group.
The new finding are still quite interesting, especially regarding the evolution of sociality within the hymenoptera, which was thought to been evolved several times (based on different phylogenies).