Doctoral Dissertations

Date of Award

5-1996

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Anthropology

Major Professor

Andrew Kramer

Committee Members

David L. Sylwester, Lyle Konigsberg, Fred Smith, Jan Simek

Abstract

The integrity of Homo habilis has been the subject of a debate that began more than thirty years ago when the species was first identified and described by Leakey, Tobias and Napier (1964). Within the collection of fossils that have been attributed to Homo habilis there is so much variation in size and shape that many paleoanthropologists believe that more than one species is represented. There have been a number of attempts to demonstrate this, but none of these has been able to conclusively demonstrate that the magnitude of variation in Homo habilis is significantly greater, or that the pattern of variation is significantly different from the pattern seen in modern hominoids. Consequently, many paleoanthropologists remain unconvinced, arguing that the variation in Homo habilis reflects nothing more than sexual dimorphism, polytypism and evolutionary change within a single lineage.

In this dissertation these competing hypotheses are evaluated by using computer- intensive randomization methods and univariate and multivariate statistics to compare variation in Homo habilis with variation in modern hominoids and fossils hominids. Randomization methods and univariate statistics are used to compare the magnitude of variation in the Homo habilis dental sample with variation in two extant species of hominoids, Pan troglodytes and Gorilla gorilla, and one fossil hominid, Australopithecus robustus. Multi- variate statistics and randomization methods are used to compare the size and shape differences between KNM-ER 1470, the largest cranium attributed to Homo habilis, and three smaller crania, KNM-ER 1805, KNM-ER 1813 and OH 24, with size and shape differences between individuals in samples of modern hominoids. For the face and vault the multivariate distances between the fossils are partitioned into 'size' and 'shape' distances whose significance is determined by comparing them with distributions of size and shape distances between individuals in reference samples representing modern Homo, Pan, Pongo and Gorilla. The multivariate distances between the Homo habilis crania are also compared with multivariate distances between other Plio-Pleistocene hominids.

The results of the univariate and multivariate comparisons strongly suggest that there is more variation in the Homo habilis cranial and dental sample than is reasonable for a single hominoid species. The dental comparisons show that the degree of relative variation in the buccolingual diameter of P4 is significantly greater than it is in any of the reference groups, even the strongly dimorphic gorillas. The pattern of variation in the early Homo dental sample also appears to be unlike the pattern seen in the reference samples, and this also suggests that more than one species is represented.

The results of the multivariate comparisons with the modern hominoids provide the strongest evidence for multiple species of early Homo. These comparisons demonstrate that in both the vault and face the size differences, and in some cases the shape differences, between KNM-ER 1470, and the smaller individuals, KNM-ER 1805, 1813 and OH 24, are large when compared to the size and shape differences seen in the extant hominoids, and it is unlikely that KNM-ER 1470 and the smaller specimens could all represent the same species, even if that species were as sexually dimorphic as Gorilla.

The size differences between KNM-ER 1470 and the smaller crania are also large when compared to the size differences between fossil hominids that are conspecific, and they are often large when compared to the size differences between fossil hominids that represent different species. The multivariate comparisons of the differences between KNM-ER 1470 and KNM-ER 1805, 1813 and OH 24 with the differences between individuals in the modern hominoids and the fossil hominids demonstrate that it is extremely unlikely that all of these specimens could belong to the same species. Thus, there were at least two lineages of gracile hominids in East Africa during the late Pliocene and early Pleistocene.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS