On bivalve phylogeny: a high-level analysis of the Bivalvia (Mollusca) based on combined morphology and DNA sequence data

Abstract. Bivalve classification has suffered in the past from the crossed-purpose discussions among paleontologists and neontologists, and many have based their proposals on single character systems. More recently, molecular biologists have investigated bivalve relationships by using only gene sequence data, ignoring paleontological and neontological data. In the present study we have compiled morphological and anatomical data with mostly new molecular evidence to provide a more stable and robust phylogenetic estimate for bivalve molluscs. The data here compiled consist of a morphological data set of 183 characters, and a molecular data set from 3 loci: 2 nuclear ribosomal genes (18S rRNA and 28S rRNA), and 1 mitochondrial coding gene (cytochrome c oxidase subunit I), totaling ∼3 Kb of sequence data for 76 molluscs (62 bivalves and 14 outgroup taxa). The data have been analyzed separately and in combination by using the direct optimization method of Wheeler (1996), and they have been evaluated under 12 analytical schemes. The combined analysis supports the monophyly of bivalves, paraphyly of protobranchiate bivalves, and monophyly of Autolamellibranchiata, Pteriomorphia, Heteroconchia, Palaeoheterodonta, and Heterodonta s.l., which includes the monophyletic taxon Anomalodesmata. These analyses strongly support the conclusion that Anomalodesmata should not receive a class status, and that the heterodont orders Myoida and Veneroida are not monophyletic. Among the most stable results of the analysis are the monophyly of Palaeoheterodonta, grouping the extant trigoniids with the freshwater unionids, and the sister-group relationship of the heterodont families Astartidae and Carditidae, which together constitute the sister taxon to the remaining heterodont bivalves. Internal relationships of the main bivalve groups are discussed on the basis of node support and clade stability.