merely belonging to the tribe, nor to the exogamous moiety of the tribe to which the woman speaking belongs, but to children of women of her generation within the totem. Can the Dieri usage extend the meaning of "child," "son," "daughter," "brother," "sister" beyond the totem to persons belonging to the same moiety of the tribe ? And is this usage found among all tribes where the totem is in full force and reckoned by descent ? If so, it affords an additional argument for the theory that the distinction between the exogamous moieties is more fundamental, and there- fore earlier, than between the totems, an additional support for the hypothesis of the intentional bisection of the horde as the begin- ning of organization.
I am not quite sure whether Dr. Howitt holds this view. Some years ago he expressed the opinion in \}c\.Q.Joiir7ial of the Anthi-opo- logical Jjistitute that the exogamous moieties were originally totem- clans. He does not repeat it in this volume, though I infer (see p. 151) that he still inclines to it. As to the origin of totems and totemism he guards himself from a definite judgement. Rightly, as it seems to me, rejecting the hypothesis that "the primary function of a totemistic group is to ensure by magic a supply of the object which gives its name to the group," as well as the hypothesis that totems originated in nicknames, he thinks more favourably of Dr. Haddon's suggestion that they arose out of the special varieties of food adopted by different groups in conse- quence of their different environment. But if so, and if the exogamous moieties were originally totem-clans, what becomes of the theory of intentional bisection? It is possible that, as he says, exogamy is merely a secondary feature of totemism, though the legends of the Alcheringa and similar traditions are a deceitful foundation for such an opinion. But, to recur to a previous query, why should there have been only two, and exactly two, such groups in every local unit called a tribe? I am puzzled there- fore ; and I would fain hope that Dr. Howitt will be good enough to tell us explicitly to what extent the Dieri and other tribes recognize the relationships of son, daughter, brother, and sister ; whether they transcend the limits of the totem-group, and if they do, how far ; and moreover, what is his opinion on the points raised in the present and the immediately preceding paragraphs.