The means and variances of dental length, breadth, and area measurements of South and East African gracile and robust australopithecines are analyzed to determine the existence of statistically significant differences due possibly to different niche utilization and divergent evolution. The study material is divided into four groups: South African gracile, South African robust, East African gracile, and East African robust.
Comparison of East and South African graciles, East African robusts and graciles, and South African robusts and graciles shows few significant F-ratios; a high frequency of significance is observed between East and South African robusts. High frequencies of significance are observed in t tests between all groups. Probit analysis, carried out on each of the four groups separately for each measurement, shows little or no significant deviation from normality; similar results are obtained when the groups are combined, suggesting the joint-normal distribution of the total australopithecine sample. High frequencies of significant t tests and low frequencies of significant F-ratios are observed when all graciles are compared with all robusts; yet few significant t tests and many significant F-ratios occur when all East African forms are combined with all South African forms.
Observed differences in dental measurements in australopithecines tend to occur on a regional rather than a morphologic basis, especially with regard to robust samples from South and East Africa. While analysis of variance and probit analysis cannot be used to establish taxonomic divisions, results suggest the inappropriateness of dental measurements in establishing an australopithecine taxonomy.