He continues by speculating on the possibility that in Australopithecus, 

 too, the upper limits of the range of cranial capacities might have ex- 

 ceeded "appreciably" a value based on the mean + 3 S.D.s. 



Still fewer supporting data are given in this statement, and it is inter- 

 esting to note that where previously he had referred only to man and 

 gorilla as having what he considered extraordinarily high upper limits he 

 now speaks of man and "the African great apes," thereby presumably in- 

 cluding chimpanzee, the other African great ape. 



Robinson thus proposes to introduce a correction factor, and, from 

 the earlier reference cited (p. 29), he bases this correction on the range 

 of variation found in gorilla and modern human cranial capacities. The 

 validity of his inference depends in large measure on 3 factors: how he has 

 derived his interracial mean and standard deviation for modern man and 

 gorilla; whether he has used appropriately weighted estimates of these 2 

 statistics; and the validity of his choice of these 2 hominoids, namely man 

 and gorilla, as the basis for analogy. 



Unfortunately, neither of the first 2 issues is dealt with, and no evi- 

 dence is cited in support of the third proposition— the validity of choosing 

 man and gorilla, but not apparently chimpanzee or orangutan. 



This last point would seem to be the most serious criticism of the 

 correction Robinson has proposed to make. Since he has made no attempt 

 to justify his choice of man and gorilla as the 2 living hominoid taxa from 

 which to derive his correction factor, it has proved necessary to examine 

 this question more closely. His proposed new australopithecine mean of 

 430 c.c. is the figure on which he has based his computations. Yet this 

 figure of 430 c.c. is far closer to the means for orangutan (411.2 c.c. for a 

 combined sample of 260 males and females) and for chimpanzee (393.8 c.c. 

 for a combined sample of 144 males and females), than it is to the means 

 for gorilla (498.3 c.c. for a combined sample of 533 males and females; or 

 506 c.c. for 653 males) or for modern man (interracial mean of 1334 c.c. 

 based on Bailey and von Bonin 1951). Despite this greater proximity of 

 his estimate of the australopithecine mean to the means for chimpanzee 

 and orangutan, Robinson has chosen the 2 taxa that, by his analysis, have 

 means further removed from his estimate for Australopithecus africanus. 

 It would seem to be implied that A. africanus is of a comparable order 

 of variability, with comparable occasional high values, to man and go- 

 rilla. 



A closer look at the distributions of the available samples of modern 



31 & 



