Phylogeny of the gastropod superfamily Cerithioidea using morphology and molecules

Abstract

The Cerithioidea is an ecologically important superfamily of basal Caenogastropoda with speciose marine, brackish water, and freshwater lineages primarily in tropical, subtropical, and warm temperate regions of the world. They often represent significant components of the communities where they occur and have given rise to several spectacular endemic radiations in rivers and ancient lakes. Earlier attempts to resolve the phylogenetic history of the group have been based on smaller taxon and character subsets with incongruent results. Here the monophyly and phylogeny of the group is evaluated with expanded morphological and molecular (16S, 28S rRNA) data sets. For morphological analyses, 151 characters (shell, operculum, radula, alimentary tract, kidney, nervous system, reproductive anatomy, and sperm ultrastructure) were scored for 47 cerithioideans (representing 17 families) and nine outgroup taxa. To test monophyly of the Cerithioidea, extended molecular data sets of 16S and 28S sequences for 57 and 44 taxa, respectively, were compiled using new and previously published sources. For combined analyses, a pruned molecular data set was combined with the morphological partition. The morphological data were analysed alone using only parsimony; molecular and simultaneous analyses were performed using both parsimony and Bayesian inference. The effect of excluding unconserved regions of the alignments was also explored. All analyses, with the exception of the individual 16S and 28S data sets, support monophyly of the Cerithioidea as currently formulated. Of the 12 families represented by more than one terminal, only two (Planaxidae, Potamididae) are always supported as monophyletic; Batillariidae, Cerithiidae, Pachychilidae, Pleuroceridae, Semisulcospiridae, Thiaridae, and Turritellidae are monophyletic in most but not all topologies. The combination of diverse data sources (morphology, 16S and 28S sequences) and inclusion of unconserved regions of the alignments improved the recovery of monophyletic families. At deeper levels, a consensus is beginning to emerge in the recognition of three main assemblages, but whether these represent clades or grades is still unclear; the resolution of these assemblages and the branching order within them are sensitive to exclusion of unconserved regions and choice of optimality criterion. No clear conclusion is reached with respect to the number of freshwater invasions, with two invasions supported on some topologies and three supported on others. Progress toward a robust and stable resolution of cerithioidean relationships will require (1) strategically coordinated sampling for additional morphological and molecular data; (2) comprehensive anatomical treatments for several poorly documented limnic lineages (e.g. Melanopsidae, Thiaridae) and comparative data for poorly understood organ systems (e.g. renal system); (3) the addition of poorly known, minute, and/or rare marine taxa, to provide novel character combinations, insight into putative homologies, and to help anchor basal nodes and break up long branches. © 2011 The Linnean Society of London.