TOTEMISM AND EXOGAMY 231 



In the first place let it be understood that by 

 most ethnologists, and, at one time, though appar- 

 ently not now, by Professor Frazer, it is held that 

 descent in the maternal line is a more archaic 

 system than that in the paternal ; indeed it is not 

 hard to understand why this should be when one 

 considers many of the phases of savage life. 



Professor Frazer, who firmly adheres to the idea 

 of a primitive system of group-marriage which 

 others refuse to admit, suggests (iv., 133) that the 

 germ of exogamy " is the deliberate bisection of 

 the whole community into two exogamous classes 

 for the purpose of preventing the sexual unions of 

 near kin." If this be correct we can imagine a 

 tribe splitting into two divisions, one called Crow 

 and the other Eagle-Hawk, and laying down the 

 law that in future Crow shall not marry Crow nor 

 Eagle-Hawk, Eagle-Hawk. It is obvious that such 

 a system would be exogamous in its character, and 

 it is notorious that such a system obtains in most, 

 though not in all, of the totemistic peoples. There 

 are, however, further complications in other cases. 

 Let us, with Professor Frazer, deal with the matter 

 by the use of letters, and let us call the two divi- 

 sions of the tribe A's and B's. Then the plan men- 

 tioned above resolves itself into the law that an A 

 must marry a B ; this is a two-class system. But 

 there is a four-class system under which A is again 

 subdivided into a 1 and a 2 and B into b l and b 2 and 

 under which also, though A must still marry B, 

 he may not marry any B, but must marry if 

 Aa l a B& 1 , if Aa 2 a Eb 2 . There is even an eight- 

 class scheme, where, of course, further prohibitions 



