A MULTIVARIATE MORPHOMETRIC ANALYSIS OF THE LATE CRETACEOUS PLACENTICERAS GENUS OF ALABAMA: SEXUAL DIMORPHISM, ONTOGENETIC VARIATION, AND A SINGLE SPECIES?

The Late Cretaceous (Santonian—?Maastrichtian) ammonites of the Placenticeras genus in the eastern Gulf Coastal Plain (USA) have historically been split into as many as two genera or subgenera (i.e., Placenticeras and Stantonoceras) and 14 species or subspecies. However, more recent work on Placenticeras in other regions has presented evidence for large amounts of intraspecific variation in this genus, suggesting that it may be taxonomically oversplit. Here, we perform morphometric analyses on a large sample (112 specimens) of Placenticeras from Alabama (and neighboring counties) to characterize the morphological variation and determine the number of distinct ‘groups’ of individuals that emerge independently from the data. Linear mixed models were used to quantify the allometric scaling between each morphometric variable and whorl height size, and to distinguish between ontogenetic (within-individual) and other intraspecific or interspecific sources of morphometric variation. Multivariate morphometric analyses including principal components analyses (PCA) and hierarchical clustering analyses were used to visualize the morphospace of our specimens and to determine the number of groups (clusters) of individuals supported by the morphometric data, which could represent intraspecific or interspecific groups (e.g., dimorphs, polymorphs, or species).

Our analyses revealed three clusters of individuals within the dataset, which we interpret as representing juveniles, microconchs, and macroconchs of a single species. Sexual dimorphism in our sample is characterized mainly by differences in relative adult size. Variables that have previously been used to distinguish between Placenticeras or Stantonoceras forms (such as relative whorl width and venter width) varied along a continuous distribution and likely represent intraspecific variation rather than clear differences between genera, subgenera, or dimorphic pairs. Our results highlight the pitfalls of the traditional typological approach to taxonomy, which often does not adequately account for intraspecific variation and results in taxonomic oversplitting. Our work also emphasizes the need for large sample sizes and quantitative methods when evaluating the taxonomy of fossil groups with large amounts of morphological variation.