Biogeography and Taxonomy of Cavellia Iredale, 1915 (Pulmonata: Charopidae)

Abstract

Cavellia is a large endemic New Zealand land snail genus comprising 56 species. Thirty-six species are described and discussed for the first time for biogeographical and morphological investigation. In order to explore ways of including morphological and geographical/geological/biogeographical information in the cladistic analysis of the characters of Cavellia species, a detailed description of the distributions of characters and species is presented. On the basis of this geographical character analysis, the standard cladistic coding procedure is modified by recoding characters to include both morphological and biogeographical information and results are compared with the standard transformational approach and three-taxon-statements (TTS). To recode the characters each state is distinguished according to the biogeographical track the species belongs to. This novel approach compared with other phylogenetic techniques including TTS achieves a better resolution of homoplasic characters and reduces homoplasy as shown by individual and ensemble tree statistics. Other phylogenetic analyses combining geological and morphological characters are also carried out. The relationships between biogeography, geology and systematics are explored using centres of diversity, track compatibility method and parsimony analysis of endemicity (PAE). Eleven centres of diversity are recognized for the genus Cavellia. When these centres are compared with the centres determined for other charopid genera, vicariance emerges as the predominant mode of taxic diversification at subfamily level as well as species level. Correlations with zones of geological activity are discussed at a regional level. Compatibility track analysis of the distributions of 23 other invertebrate genera and subgenera provides some support for the tracks identified for Cavellia species by character geography. Of sixty-eight identified tracks, eight are replicated at least in one other invertebrate group. Many more distributions need to be analysed in order to derive a sound classification of tracks, which can then be fed back into the phylogenetic analysis. Some baselines or diagnostic characters have been suggested for the Cavellia tracks but still await comparison with similar existing studies. PAE has been criticised for ignoring taxonomic relationships by using solely distributional data. However, when the presence and absence of species from geological areas is analysed by PAE, the resulting clusters conform to regions that are recognized by geologists as sharing a common history. Similarly, when the presence and absence of geological characters from areas is analysed by PAE, areas are clustered together that agree with the geological history of New Zealand. Until now the concept of biological homology has not been used in cladistic analysis but further investigation into ways of including biogeographical information is still needed to produce better results.