has been paid.”[242] Notwithstanding the example of the Todas, it must not be thought that a scarcity of women is essential to the existence of polyandry. In Tibet this system of marriage is universal, and it has been so from time immemorial. Nevertheless, unmarried women are numerous, and infanticide is not practised. Mr. Andrew Wilson defined Tibetan polyandry as the marriage of one woman to two or more brothers, and these are actual brothers, although at one time probably they may also have been tribal. The choice of a wife is the right of the elder brother, and Mr. Wilson states[243] that “among the Tibetan-speaking people it universally prevails that the contract he makes is understood to involve a marital contract with all the brothers, if they choose to avail themselves of it.” Moreover, all the children of the marriage belong to the eldest brother, as the head of the family group. In Ladak,[244] however, the consent of the younger brothers is required to the marital partnership, although on the death of the eldest brother his authority, with his property and his widow, devolve upon his next brother, whether or not there has been a polyandrous arrangement. Mr. Wilson observes[245] that Tibetan polyandry had the effect “of checking the increase of population in regions from which emigration is difficult, and where it is also difficult to increase the means of subsistence.” It is due to an artificial scarcity of _wives_, rather than of women, in which it differs from the polyandry of the Todas, which is the consequence of an actual scarcity of females, caused originally by the practice of infanticide, and afterwards by a preponderance of male births. Both the Tibetans and the Todas trace descent through the male line—that is, take the family or gentile name of the father; but some peoples of Southern India, who practice polyandry, prefer the female line. This is not surprising, when we find, as among the Nairs of Malabar, that not only has a woman several husbands, but a man “may be one in several combinations of husbands.” Such unions, which are governed by certain restrictions as to tribe and caste, closely resemble the Australian group marriage. In Ceylon, where polyandry is very prevalent among the Kandyans, marriage is of two forms, one termed _deega_, in which the wife goes to live in the house and village of her husband or husbands, the other, termed _beena_, in which the husband or husbands come to reside with her in the house of her birth. The Tibetan polyandry may be a form of the _deega_ marriage, and the Nair polyandry a form of the _beena_ marriage, although it is possible that the latter may be a “mere freak,” if it be true (as Mr. Wilson affirms) that the Nairs are nominally married to girls of their own caste, but never have any intercourse with their wives, who may have as many lovers as they please, provided they are Brahmins or Nairs, other than the husband. These lovers answer to the