Distinguishing between two Antarctic species of Eocene Palaeeudyptes penguins: A statistical approach using tarsometatarsi

Abstract

Defining species boundaries, due to morphological variation, often represents a significant challenge in paleozoology. In this paper we report results from multi- and univariate data analyses, such as enhanced clustering techniques, principal coordinates ordination method, kernel density estimations and finite mixture model analyses, revealing some morphometric patterns within the Eocene Antarctic representatives of Palaeeudyptes penguins. These large-sized birds were represented by two species, P. gunnari and P. klekowskii, known mainly from numerous isolated bones. Investigations focused on tarsometatarsi, crucial bones in paleontology of early penguins, resulted in a probability-based framework allowing for the "fuzzy" partitioning the studied specimens into two taxa with partly overlapping size distributions. Such a number of species was supported by outcomes from both multi- and univariate studies. In our opinion, more reliance should be placed on the quantitative analysis of form when distinguishing between species within the Antarctic Palaeeudyptes.