Craniometric variation in large-bodied hominoids: testing the single-species hypothesis for Homo habilis

Abstract

The species question regarding Homo habilis remains unresolved. This study tests the null hypothesis that this hominid taxon represents a single paleospecies by comparing the fossils' magnitude and pattern of craniometric variation to samples of extant, large-bodied hominoids including Pan troglodytes, Gorilla gorilla, Pongo pygmaeus and Homo sapiens. Unlike previous quantitative approaches to this problem, randomization methodology is used to assess the significance of univariate and multivariate measures of variation. The benefits of randomization include its applicability to problems of small sample size, its robusticity in producing statistical probabilities, and its freedom from underlying assumptions concerning the normality of variable distribution. This study demonstrates that the magnitude of variation in H. habilis does not exceed that displayed by the most sexually dimorphic apes such as orang-utans and gorillas, but that the pattern of variation in these early Homo fossils is unlike that shared by any of the modern referents. Although the effects of geological time are considered as possibly responsible for the unique pattern of craniometric variation seen in H. habilis , we interpret these results as providing tentative support for a multiple-species solution, thereby provisionally rejecting the null hypothesis of Homo habilis as a single species.