the same time, the vertical dimension is increased . . . the loss of area from con- 

 striction of the breadth is almost compensated by the increase in depth. The 

 [cross-sectional] area remains essentially constant, and since the length of the 

 endocast is not affected by constriction the volume remains essentially constant 

 also. [Holloway 1965, p. 205] 



To arrive at this conclusion, Holloway varied the angle along the 

 sagittal suture far more than necessary, that is, he increased the maximum 

 (biasterionic) breadth by as much as 10 mm. and reduced it by as much 

 as 20 mm. This is an extreme range, compared with the 2 mm. variation 

 we found between our 2 reconstructions: despite his deliberate choice of 

 what he describes as a "range surely greater than necessary," he still showed 

 relative constancy of the volume determined. Thus, he concluded that my 

 values for the partial endocranial volume were probably correct, provided 

 that the reconstruction of the 2 parietal bones was valid. 



Correcting from partial endocast to the total cranial capacity. 

 And now the problem was to estimate what proportion of the total endo- 

 cranial volume might be made up by the biparietal endocranial volume. In 

 proceeding to this next step, I was simply following a long-recognized pro- 

 cedure in palaeoanthropology for the estimate of total from partial endo- 

 cranial capacity— such, for example, as had been used by Dubois, Schaaffhau- 

 sen, Schwalbe, T. H. Huxley, Sollas, and Weidenreich. What was new about 

 this particular exercise, as far as I have been able to determine, was the 

 use of so unusual a portion of the cranial cavity as the biparietal tunnel. 

 Previous studies have tended to use a calotte without the calvarial base— 

 for example Dubois's (1898) determination of the capacity of the Trinil 

 calvaria. 



To provide a basis for making my estimate, I selected 4 endocranial 

 casts of fossil hominids in which the total volume was fairly reliably 

 known and on which the outline of the parietal bones was unmistakeable. 

 Two were endocasts of australopithecines (the Taung specimen and Old. 

 Horn. 5 or A. boisei from Olduvai) and 2 of Homo erectus (1 each from 

 Indonesia and China). From each of the 4 endocasts biparietal partial casts 

 were prepared and the volumes determined by water displacement 5 times 

 each. For each cast the mean values obtained were then expressed as a 

 percentage of the known total volume. 



The percentage for the Taung specimen came to 56.54; that for 

 A. boisei (Old. Horn. 5) came to 54.45 (Figure 18); and that for H. erectus 

 pekinensis came to 54.64 (Figure 19). The value for the Trinil specimen, 



* 66 



