Carboniferous non-marine bivalves are usually classified at the ‘species’ level using external morphological criteria through comparison with figured and described type specimens. This system of nomenclature has allowed the recognition of stratigraphically useful morphospecies.
Biometric analysis of a large number of shells of the genus Anthraconauta Pruvost, collected from the Sydney coalfield, Nova Scotia, suggests. that this system of nomenclature cannot be justified biologically. The genus appears to be morphologically homogeneous in the Upper Carboniferous of Western Europe and North America. Shell shape variation is influenced by mode of life and palaeoenvironment, as suggested by the study of modern analogues and fossil shells. If a biological classification were adopted for this group of fossils, then all their stratigraphical value would be lost.
To reconcile systematics with the needs of stratigraphers, the adoption of a trinomial nomenclature is recommended. The use of cluster analysis has shown that varieties can be identified that correspond morphologically to previously named and described ‘species’. Nomenclature of the form genus, species and variety would be both biologically correct and stratigraphically useful.