Species names and metaphyly: a case study in Discodorididae (Mollusca, Gastropoda, Euthyneura, Nudibranchia, Doridina)

Dayrat, B. & Gosliner, T. M. (2005). Species names and metaphyly: a case study in Discodorididae (Mollusca, Gastropoda, Euthyneura, Nudibranchia, Doridina) —Zoologica Scripta, **, ***–***. Absence of resolution in phylogenetic trees, or metaphyly, is a common phenomenon. It mainly results from the fact that each data set has its own limit and can hardly be expected to reconstruct alone an entire hierarchy. Because metaphyly helps point out which regions of a tree merit further investigation, one should not avoid metaphyly but rather should try to detect it by addressing carefully node reliability. In this paper we explore the implication of metaphyly for species names. We present a phylogenetic analysis of Discodorididae (Mollusca, Gastropoda, Nudibranchia, Doridina), with an emphasis on relationships among species of Discodoris and its traditionally close ‘allies’ such as Peltodoris and Anisodoris. We demonstrate that some species must be transferred to different discodoridid subclades with which they share synapomorphies, but that many species form a metaphyletic group at the base of Discodorididae, and therefore cannot be placed in any taxon of genus level. We demonstrate that the current International Code of Zoological Nomenclature does not allow taxonomists to handle this situation because it requires selecting a taxon name of genus rank for every species binomial. Then we evaluate the results provided by new forms of species names, both in a rank‐based system, such as the current nomenclature, and a rank‐free system. All solutions considered would cause radical changes to the ‘spirit’ of the current ICZN (and, by extension, to the other current codes). In a rank‐free system of nomenclature, such as the PhyloCode, the best result is obtained with an epithet‐based form that does not mention supra‐specific relationships. Under this method, official species names would take the form ‘boholiensis Bergh, 1877’, although page numbers and letters can be added for uniqueness purposes. Taxonomists would then be free to add supra‐specific taxon names in ‘common’ species names, such as Discodorididae boholiensis Bergh, 1877 or simply Discodorididae boholiensis. Here we wish to stimulate discussion of a problem that we believe merits wide debate: absence of resolution in phylogenetic reconstruction and its impact on species nomenclature.