The other social forms of sexual intercourse, forms whose existence is admitted even by the critics of promiscuity, are characterized by frequent =changes= in sexual relationships. This is especially true of the oldest form of marriage, the so-called “=group-marriage=.”[160] Group-marriage is not a union in marriage of isolated individuals, but such a union between two =tribal groups=, composed respectively of male and female individuals, a union between the so-called =totems=. The social instinct, the impulse towards companionship, upon which even to-day the State and the family depend, united mankind at one time into tribes of a peculiar kind, which felt themselves to constitute single individuals, and believed themselves to be inspired by an animal spirit, their protective spirit. Their union was known as the totem. Group-marriage is =the marriage of one totem with another=--that is, the men of one totem-group marry the women of another, and _vice versa_. But =no individual man has any particular wife=. On the contrary, if, for example, twenty men of the first totem espoused twenty women of the second totem, then each one of the twenty men had an equivalent share of each one of the twenty women, and _vice versa_. This was indeed an advance over unrestricted sexual promiscuity, limited by no social forms; but it afforded no possibility of any individual relationships of love, it remained promiscuity within narrow bounds. Group-marriages exist at the present day in Australia in a well-developed form among certain tribes; whilst, as an occasional custom, in the form of an exchange of wives among friends, guests, and relatives, it appears to be almost universally diffused throughout Australia. Schurtz regards Australian group-marriage as a kind of partial taming of the wild sexual impulse. Well known is the description of group-marriage in ancient Britain given by Julius Cæsar: “The husbands possess their wives to the number of ten or twelve in common, and more especially brothers with brothers, or parents with children.” Here we have a special variety of group-marriage. According to Bernhöft, =polyandry= is also to be regarded as the vestige of a primitive form of group-marriage, arising from a deficiency of women in a totem, so that one woman was left as the representative of the totem married to several husbands. Marshall has, in fact, amongst the polyandrous Toda in Southern India, actually observed group-marriage side by side with polyandry. Among certain Indian tribes we find even at the present day indications of group-marriage. For example, the husband will have a claim on the sisters of his wife, or even on her cousins or her aunts, and gradually he may marry them. In this case we see that =polygyny= has developed out of the group-marriage.