Variation among the Dmanisi hominins: Multiple taxa or one species?

Abstract

There is continuing controversy over the number of taxa documented by the Dmanisi hominins. Variation may reflect age and sex differences within a single population. Alternatively, two (or more) distinct species may be present. Our null hypothesis states that just one population is represented at the site. We assess the likely sources of variation in endocranial capacity, craniofacial and mandibular morphology, and the expression of characters related to aging and sex dimorphism. We use the coefficient of variation and a modified version of Levene's test for equal variances to compare trait variation at Dmanisi with that in fossil hominins and modern Homo sapiens from Africa. Skull 5 presents a low, massive vault, and a muzzle-like lower face. Other individuals have larger brains and more globular vaults. Despite such variation, the five crania share numerous features. All of the mandibles possess marginal tubercles, mandibular tori, and a distinctive patterning of mental foramina. Relative variation at Dmanisi is comparable to that in selected reference groups. Further growth anticipated in Skull 3, age-related remodeling affecting the D2600 mandible, pathology, and sex dimorphism can account for much of the interindividual variation observed. The preponderance of evidence supports our null hypothesis. Sources of the variation within ancient Homo assemblages remain poorly understood. Skull 5 is a very robust male, with a brain smaller than that of both a juvenile (Skull 3) and a probable female (Skull 2). Skull 1 has the largest brain, but cranial superstructures do not clearly mark this individual as male or female. It is likely that the Dmanisi hominins represent a single paleospecies of Homo displaying a pattern of sex dimorphism not seen in living hominids.