At all events, the Arunta ‘clan’ (totem kin) is now no longer exogamous, and two Arunta phratries can now contain members of the same totems, contrary to Kamilaroi, Dieri and Urabunna and American custom. How did this anomaly arise? Dr. Durkheim supposes that the change began when Arunta kinship came to desert the female and to be reckoned in the male line. This appears to Dr. Durkheim to be indicated by the complicated and ingenious arrangements made when an Urabunna (who reckons by the female line) intermarries with an Arunta, who reckons by the male line.[4] These arrangements, he thinks, are no novelty devised for the occasion: the Arunta merely revert to their old way of reckoning by the spindle side. When the Arunta changed their system, and reckoned in the male, not, as of old, in the female line, the children now belonged to the ‘phratries,’ not of their mothers, as previously, but of their fathers. Each ‘phratry’ then bartered a sub-class of its own for a sub-class of its partner. Each bartered sub-class thus brought its totems into the other ‘phratry,’ and there was no longer a totem group entirely peculiar to one or other ‘phratry.’ Consequently, a member of the Kangaroo totem could marry a woman of the same, if she were in the opposite ‘phratry’ to his own. Might not the same results follow from the mere fact, that, among the Arunta, the totem is now inherited neither from father nor mother, but is derived simply from the totem souls that haunt the particular glen or hill where the child was conceived? By this means a totem soul can get into a child of the ‘phratry’ to which that totem did not originally belong, and thus the totems ‘skip’ from one ‘phratry’ to another, contrary to general rule in Australia and North America. This is the explanation of the Arunta anomaly which Messrs. Spencer and Gillen accept. ‘The spirit child’ (of the Lizard totem) ‘deliberately, the natives say, chose to go into a Kumura’ (class) ‘woman, instead of a Bulthara woman.... Though the class was changed, the totem could not possibly be.... Owing to the system according to which totem names are acquired, it is always possible for a man to be, say, a Purula’ (class) ‘or a Kumura’ (class) ‘and yet a Witchetty; or, on the other hand, a Bulthara’ (class) ‘or a Panunga’ (class) ‘and yet an Emu’ (totem). But, if he is thus born to a totem which was not originally (on my theory) a totem of his phratry, a man loses the chance of being an _Alatunja_, or head man of a local group.[5] Thus the Arunta anomaly arises merely and necessarily from the Arunta philosophy of souls. That philosophy is an isolated freak, and it has upset and revolutionised Arunta Totemism, which, therefore, is the reverse of the ‘primitive’ model. [Footnote 1: _L’Année Sociologique_, v. 91, 92.] [Footnote 2: This idea we shall find again later, in another part of Dr. Durkheim’s system.] [Footnote 3: _L’Année Sociologique_, i. 6, 7.]