In the preceding pages I have collected together, with as much exactitude as I could, all that we know of the familial clan and the maternal family. It would be bold to assert that such have been the primitive and always necessary forms of the family. It is indisputable, however, that they are or have been very common in all countries and in all races. But these are types of familial association already regulated and complicated. Anterior to them there must have existed, in the little human hordes, a complete anarchy, most often characterised by the despotism of the strongest male, dominating a small flock of women and children, who were meekly submissive to his caprices--in fact, a sort of bestial patriarchate. Among thinly-scattered races, without intelligence or industry, practical monogamy was established at the very outset. We know that this was the case with the stupid Veddahs of Ceylon, when they wandered in simple families in the virgin forests of their island, incapable of constituting even the smallest horde. As soon as men have grouped themselves in small societies, regulated even in a slight degree, the familial clan with its confused kinship must frequently have been constituted, but on plans which were necessarily variable according to the conditions and exigencies of the social life. All that was possible has surely been attempted; sometimes regulated promiscuity, for each man claimed his rights, sometimes the mixed polyandric and polygynic household, elsewhere simple polyandry, when the women were scarce, and at other times monogamy. I repeat, all that was compatible with the maintenance of the little social group must have been tried at first; and then selection assured the permanent adoption of such or such a system. As soon as men began to take note of descent, it was always uterine filiation that they held in account; paternal descent was less evident, and less easy to prove; it has been nearly everywhere the latest, and the widely-spread custom of the couvade proves that it was not established without difficulty. It has ended, however, by triumphing, all over the world, in the states that are still barbarous, but which have a complex social and political structure, where the primitive tribal _régime_ has more or less disappeared, and where a line of demarcation sufficiently strong separates the interests of the individual from those of the group to which he belongs. In short, the social transformation from which the paternal family has arisen has nearly always coincided with a radical change in the _régime_ of property, which has simultaneously become individual or, at least, familial. FOOTNOTES: [976] Layland, _Journal of Ethnological Society_, 1869. [977] Bosman, _Voyage en Guinée_, p. 197. [978] Bowdich, _Observations sur le Gouvernement des Achantis_ (collection Walkenær, t. xii.). [979] A. Giraud-Teulon, _Orig. de la Famille_, p. 216. [980] Herbert Spencer, _Sociology_, vol. ii. p. 340.