Polyandry, or the union of one woman with several men, is a relatively rare form of marriage and the family, found only in certain isolated regions of the world. It is particularly found in Tibet, a barren and inhospitable plateau north of India and forming a part of the Chinese Empire. It is also found in certain other isolated mountainous regions in India, and down to recent times also in Arabia. In none of these places does it exist exclusively, but rather alongside of monogamy and perhaps other forms of the family. Thus in Tibet the upper classes practice polygyny and monogamy, while among the lower classes we find polyandry and monogamy. In all these regions where polyandry occurs, moreover, it is to be noted that the conditions of life are harsh and severe. Tibet is an exceptionally inhospitable region, with a climate of arctic rigor, the people living mainly by grazing. Under such circumstances it is conceivably difficult for one man to support and protect a family. At any rate, the form of polyandry which we find in Tibet suggests that such economic conditions may have been the main cause of its existence. Ordinarily in Tibet a polyandrous family is formed by an older brother taking a wife, and then admitting his younger brothers into partnership with him. The older brother is frequently absent from home, looking after the flocks, and in his absence one of the younger brothers assumes the headship of the family. Under such circumstances we can see how the natural human instincts which would oppose polyandry under ordinary circumstances, namely, the jealousy of the male, might become greatly modified, or cease to act altogether. Certain other conditions besides economic ones might also favor the existence of polyandry, such as the scarcity of women. Summing up, we can say, then, that this rare form of the family seems to have as its causes: (1) In barren and inhospitable countries the labor of one man is sometimes found not sufficient to support a family. (2) Also there probably exists in such regions an excess of males. This might be due to one of two causes: First, the practice of exposing female infants might lead to a scarcity of women; secondly, in such regions it is found that from causes not well understood a larger number of males are born. It may be noted as a general fact that when the conditions of life are hard in human society, owing to famine, war, or barrenness of the soil, a larger number of male births take place. We may therefore infer that this would disturb the numerical proportion of the sexes in such regions. (3) A third cause may be suggested as having something to do with the matter, namely, that habits of close inbreeding, or intermarriage, might perhaps tend to overcome the natural repugnance to such a relation. Moreover, close inbreeding also, as the experiments of stock-breeders show, would tend to produce a surplus of male births, and so would act finally in the same way as the second cause.