with my former estimate of 461 to 643 c.c. for Australopithecus sensu lato— 

 Tobias 1968b— now hereby superseded: and with Robinson's estimate of 300 

 to 550 c.c. for A. africanus— Robinson 1966.) My latest estimate of the A. 

 africanus population range is based upon a distribution about the present 

 sample mean of 494 c.c. This would be a valid estimate if the present 

 sample of 6 A. africanus capacities were drawn from the middle reaches of 

 the population distribution. Such values as 650 c.c. for the braincase repre- 

 sented by the Makapansgat occipital would then be improbable. 



However, through sampling bias, the present australopithecine sample 

 may give a poor indication of the population mean. In the extreme case, if 

 540 c.c, the value that was estimated for the Taung adult and that forms 

 the highest value in the sample, were also the highest value in the popula- 

 tion, the estimated population range would be 291 to 540 c.c. Conversely, if 

 435 c.c, the lowest value in the sample, coincided with the lowest value in 

 the population, the estimated population range would be 435 to 684 c.c. 



It may tentatively be concluded that if the 6 australopithecines whose 

 cranial capacities have been enumerated above were members of a single 

 population, and if the population variability (as expressed by the range) 

 were of the order of. 6 times the estimated standard deviation, then it would 

 be highly unlikely that any normal adult A. africanus cranium still to be 

 discovered would have a capacity smaller than 291 c.c. or greater than 684 

 c.c* The latter value is smaller than the smallest capacity in H. erectus 

 (750 c.c). It is smaller than 2 earlier estimates of an imaginary cerebral 

 "Rubicon" between ape and man, that of Vallois (800 c.c.) and that of Keith 

 (750 c.c). Interestingly, too, the estimated upper value of 684 c.c is exceeded 

 by the largest capacity known in a gorilla (752 c.c.)! 



We may conclude that the estimates of 1000 c.c. for the upper limits 

 of the australopithecine cranial capacity (Broom and Robinson 1952; Dart 

 1956a) are excessive and highly unlikely. In a former estimate I stated that 

 "the most generous estimate yields a maximum of 848 c.c. at the australo- 

 pithecine grade of hominid organization" (Tobias 1963); but this value, 

 which is now seen to be far too high, was based on comparison with living 

 hominoids and not on a statistical analysis of the variance of the sample 

 itself. The reason that comparison with values for living hominoids yields 

 improbable results is discussed below. 



The "middle reach" range of 370 to 618 c.c. is the most likely popula- 



* On the basis of Holloway's (1970b) reestimates. these extreme outside limits would be of the 

 order of 350 to 560 c.c. 



^ 28 



