that with pastoral peoples the like occurs, we have direct evidence: Pallas tells us of the Calmucks and Mongols that men oppressed by a chief desert and go over to other chiefs. Occasionally occurring everywhere, this fleeing from tribe to tribe entails ceremonies of incorporation if the stranger is of fit rank and worth—exchange of names, mingling of portions of blood, etc.—by which he is supposed to be made one in nature with those he has joined. What happens when the group, instead of being of the hunting type, is of the patriarchal type? Adoption into the tribe now becomes adoption into the family. The two being one—the family being otherwise called, as in Hebrew, "the tent"—political incorporation is the same as domestic incorporation. And adoption into the family, thus established as a sequence of primitive adoption into the tribe, long persists in the derived societies when its original meaning is lost.
And now to test this interpretation. Distinct in nature as are sundry races leading pastoral lives, we find that they have evolved this social type when subject to these particular conditions. That it was the type among early Semites does not need saying: they, in fact, having largely served to exemplify its traits. That the Aryans during their nomadic stage displayed it is implied by the account given above of Sir Henry Maine's investigations and inferences. We find it again among the Mongolian peoples of Asia; and again among wholly alien peoples inhabiting South Africa. Of the Hottentots, who, exclusively pastoral, differ from the neighboring Bechuanas and Caffres in not cultivating the soil at all, we learn from Kolben that all estates "descend to the eldest son, or, where a son is wanting, to the next male relation;" and "an eldest son may after his father's death retain his brothers and sisters in a sort of slavery." Let us note, too, that among the neighboring Damaras, who, also exclusively pastoral, are unlike in the respect that kinship in the female line still partially survives, patriarchal organization, whether of the family or the tribe, is but little developed, and the subordination small; and further, that among the Caffres, who, though in large measure pastoral, are partly agricultural, patriarchal rule, private and public, is qualified.
It would doubtless be unsafe to say that under no other conditions than those furnished by pastoral life does there arise this family type. We have no proof that it may not arise along with direct transition from the hunting life to the agricultural life. But it would appear that usually this direct transition is accompanied by a different set of changes. Where, as in Polynesia, pastoral life has been impossible, or where, as in Peru and Mexico, we have no reason to suppose that it ever existed, the political and domestic arrangements, still characterized much or little by the primitive system of descent in the female line, have acquired qualified forms of male descent and its concomitant arrangements; but they appear to have done so under pressure of the influences which habitual militancy maintains. We have an indication of
