162 WARFARE IN THE HUMAN BODY 



pology, must be more or less familiar with the facts of 

 cannibalism, there is no need to enter into a prolonged 

 enumeration of its phenomena which are not yet co- 

 ordinated. In preference to otiose repetition I shall 

 therefore examine a few of the more remarkable customs 

 connected with anthropophagism, for, if any solution can 

 be obtained of them, many, if not all, of the remaining 

 details will fall into their place automatically. Premising, 

 then, that it is admitted there is sufficient evidence to 

 assume provisionally that cannibalism is a stage through 

 which early mankind has passed, it may be asked how it is 

 that in certain Australian tribes, such as the Binbinga, 

 the two classes eat, not their own, but each other's dead. 

 In many other cases, according to Spencer and Gillen, it is 

 suspected that the same custom obtains. It seems, how- 

 ever, as might have been prophesied, that those who are 

 eaten are never, or very rarely, of the same totems as those 

 who eat them. Such a custom is totally unintelligible on 

 the division theory, and remains, as it were, a mere morbid 

 degeneration such as some see in all cannibalism. Although 

 many variations in man-eating must inevitably occur as 

 totemism and tribal organization decay, what is to be 

 sought is the origin of this peculiar custom, and if any 

 hypothesis suggests it, the explanation should have solid 

 foundations. 



If the original patriarchal family was such, or nearly 

 such, as Atkinson pictured it, and as I myself drew it, it 

 can be seen how the single one-class, and single-totemed, 

 group came into being. As said before, there is no need 

 to follow Atkinson in his hypothesis that maternal love 

 overcame the hostility of the brutal father and chief, for 

 in such times and conditions as those which made the 

 environment of nascent mankind, it is safer to infer that 



