Abstract: The thickness of the cranial vault at the midline on the mid-frontal squama, pre-bregmatic eminence, frontal at bregma, parietal at vertex, occipital at lambda and the external occipital protuberance was recorded in 40 male and 7 female Northern Chinese crania, 47 male and 52 female Australian Aboriginal crania and 13 male European crania using specially nodified vernier calipers. Comparison of vault thickness data obtained through direct measurement with those obtained from lateral radiographs indicated that direct measurenent provided consistently more accurate results.
Male and female samples were processed separately so that the extent of sexbased variation could be examined. Student's t test was used to compare the sample means and the percentage of sexual dimorphism for each dimension was calculated according to Garn et al. (1964) . The possibility of an allometric association between the thickness of the bones within the cranial vault, size of the cranial vault and stature was examined using Spearman's rank correlation coefficient and the Australian Aboriginal sample.
All but one of the mean thickness dirmensions in the Australian Aboriginal male sample is significantly greater than the Northern Chinese and European means. The female results support those obtained with the males. In both males and females thickness at the external occipital protuberance, in all of the populations examined, did not correlate highly with that obtained from other parts of the cranial vault. This reflects the high degree of morphological variation in the position of the internal occipital protuberance and its influence on cranial vault thickness dimensions recorded at the external occipital protuberance. The European and Northern Chinese samples have similar cranial vault thickness dimensions. The Spearman's rank correlation coefficient matrix scores provide some support for a biological association between vault thickness and overall cranial size. However, there appears to be little support for an association between stature and cranial vault thickness. The difference between the male and female mean vault thickness dimensions were significant at bregma, vertex and the external occipital protuberance in Australian Aboriginals and lambda and the external occipital protuberance in Northern Chinese. Some caution is needed in the interpretation of the Northern Chinese female data as the sample is extremely small.
Evidence of trauma, supressed fractures, is extremely common on the vaults of Australian Aboriginal crania from southern and central Australia. Traditionally Australian Aboriginals, males and females, involved in agressive dispute will use a substantial wooden implement and strike to the head of thir opponent (Meggitt 1962) . The injuries that result from this are more common in females than in male. This form of social interaction must have rigorously selected against those individuals with thinner bones in their cranial vaults. To a large degree this may explain the greatly thickened vaults in Australian Aboriginals relative to Europeans and Northern Chines. This may also provide a clue to the factors resulting in the development of marked cranial vault thickness in Homo erecius.

Key words: Cranial vault thickness, Homo Sapiens, Northern Chinese, European, Australian Aboriginal