been assigned as type and paratypes of the larger-brained group, which 

 Leakey, Tobias, and Napier (1964) have called Homo habilis. 



Seven of these 8 australopithecine specimens are from South Africa— 4 

 from Sterkfontein, 1 from Taung, 1 from Makapansgat, and 1 from Swart- 

 krans— whilst the eighth specimen is from Olduvai. Those from Sterk- 

 fontein, Taung, and Makapansgat are generally regarded as belonging to 

 the gracile species, A. africanus (Robinson 1956), though Tobias (1967a, 

 1968a) has drawn attention to certain features of the Makapansgat specimens 

 that suggest a degree of intermediacy between the gracile and robust forms. 

 The specimen from Swartkrans is of a robust australopithecine, A. robustus 

 (sometimes regarded as a separate genus, Paranthropus); whilst that from 

 Olduvai is the type specimen of the hyper-robust species, A. boisei (Tobias 

 1967a). Thus, we have 6 estimates of cranial capacity for A. africanus, and 

 1 each for A. robustus and A. boisei. Let us consider the individual data. 



The gracile australopithecines 



The Taung child skull. This beautifully preserved skull is the type 

 specimen of A. africanus (Dart 1925). It includes a well-preserved natural 

 endocranial cast (Figure 1). The endocast comprises the impression of 

 virtually all of the right cerebral hemisphere, though the frontal pole and 

 adjacent rostral region are broken off from the main body of the endocast 

 and remain embedded against the frontal bone and the anterior cranial 

 fossa. Posteriorly, the endocast stops short of the midline, the plane of 

 cleavage diverging from the median sagittal plane and approaching close to 

 the impression of the right cerebellar hemisphere. Thus, a sufficient amount 

 of the endocast is preserved to permit a fair reconstruction to be made of 

 the hemicast and of the entire cast. 



In his original paper, it is of interest to note that Professor Dart gave 

 no estimate of the cranial capacity; he gave an estimate of the length of the 

 cranial cavity ("could not have been less than 114 mm.") and made certain 

 comparisons with the cranial cavity of a chimpanzee skull and with a large 

 endocast of a gorilla. From these statements, Sir Arthur Keith made a rough 

 estimate of the approximate "brain volume" (i.e., endocranial capacity), 

 namely, "less than 450 c.c." (Keith 1925, p. 234). 



In his 1926 paper in Natural History, Dart stated that the endocranial 

 volume was 520 c.c. The reconstruction of the hemi-endocast on which this 

 computation was based is still present in the Anatomy Department of the 



^ '2 



