Utility of complete large and small subunit rRNA genes in resolving the phylogeny of the Neodermata (Platyhelminthes): implications and a review of the cercomer theory

We combined nearly complete sequences of large (LSU) and small (SSU) subunit rDNA from 32 flatworm species to estimate the phylogeny of the Platyhelminthes using maximum parsimony, maximum likelihood and Bayesian inference methods. Rooted against the Catenulida, combined evidence trees offered no support for the Revertospermata, which was also rejected by constraint analysis. Generally, nodal support was higher for groupings estimated from the combined data partitions and all methods of analysis provided congruent estimates of phylogeny. The Monogenea and Proseriata were resolved as monophyletic, rejecting previous suggestions of paraphyly based on SSU and partial LSU data sets and thus supporting widely accepted morphological synapomorphies. Monophyly of the Neodermata was supported and its sister group was a clade of neoophoran turbellarians' to the exclusion of the Proseriata which in turn was more basal. Taxa with similar spermatology to the Neodermata (Ichthyophaga, Notentera, Urastoma and Kronborgia) were the sister group to Tricladida + Prolecithophora, which in turn were sister to the Rhabdocoela. Polycladida + Macrostomida + Lecithoepitheliata was the earliest divergent offshoot of the Rhabditophora. Among the Neodermata, the Cercomeromorphae (Cestoda + Monogenea) was not supported, whereas Cestoda + Trematoda was well supported. Although there is no known synapomorphy for this latter grouping, our data highlight problems associated with the 'cercomer theory' and we reject putative homologies regarding neodermatan 'cercomers' that have been sustained in the literature without careful scrutiny. (C) 2003 The Linnean Society of London.