Specimens of the oyster genus Cubitostrea from lower and middle Eocene beds in Maryland and Virginia were compared with collections of that taxon from correlative units in North Carolina, South Carolina, Alabama, and Mississippi. Material from these collections indicates some differences with Stenzel’s model of successional speciation. Stenzel proposed a single lineage of Cubitostrea including (from oldest to youngest) C. perplicata (Dall)—1.m. Eocene, C. lisbonensis (Harris)—m.m. Eocene, C. smithvillensis (Harris)—m.m. Eocene, and C. sellaeformis (Conrad)—m.m. Eocene. Stenzel’s model suggested a change from a primitive trigonal form to one with well-developed auricles and finally to one with a profound fold in the valves. The development of auricles and folds in the genus Cubitostrea appears to depend on the maturity of the individuals, however, and is not an evolutionary progression, because the oldest known representative, Cubitostrea sp. from the lower Eocene Nanjemoy Formation, exhibits both of these tendencies.
We believe that there are two distinct oyster lineages, both belonging to the genus Cubitostrea. The first lineage consists of C. perplicata and C. lisbonensis; the second consists of C. sp. and C. sellaeformis. C. smithvillensis appears to be an ecophenotypic variant of C. sellaeformis as are C. divaricata (Lea) and C. vermilla (de Gregorio). Specimens of Cubitostrea that have been identified as C. lisbonensis from South Carolina (Santee Limestone) and North Carolina (Castle Hayne formation) are immature and resemble the young of C. sellaeformis from Virginia and Alabama.