Nevertheless, human societies, small or large, must and will live, and it is an imperious condition which imposes the polyandric _régime_, namely, a considerable inequality between the number of men and that of women. Now, this disproportion may result from divers causes. In the first place, it may be natural, as it is among certain animal species. Among the lepidoptera, for example, nine hundred and thirty-four males have been counted as against seven hundred and sixty-one females.[169] Although smaller, the disproportion is not less real amongst mankind. As a general rule, and in nearly all countries where it has been possible to ascertain it, the relation between masculine and feminine births gives a certain excess of boys. This relation has been found in Europe to be 106 for 70 million of births; but our great masculine mortality re-establishes the equilibrium in the early years of life. The proportion of births, besides, is far from being identical in all the countries of Europe, and we even find oscillations in a given country. In England it is generally 104.5, in France 106.3, in Russia 108.9, at Philadelphia 110.5. In certain ethnic or social categories the proportion of masculine births notably augments. It rises to 113 for the Jews of Russia, to 114 for those of Breslau, to 120 for the Jews of Livonia. More singular still, the proportion of masculine to feminine births augments for certain professions; it is, for example, higher amongst the English clergy.[170] It is even seen to vary spontaneously. In the year 1886, during several months, the proportion of feminine births rose at Paris. In France, for a period of forty-four years, it has happened five times in one department, and six times in another, that the female births have been in excess. At the Cape of Good Hope, among the whites, for several years there have been ninety to ninety-nine masculine births. The reason or reasons of these spontaneous oscillations in the proportion of the sexes still escapes us. We verify it only, and we are warranted in concluding that the production of sex in the embryo depends on some relatively second causes. It is sure, for example, that the clergy of England are not of a special race. If, however, they have more male children than the other inhabitants of England, the fact can only depend on intimate particulars of their kind of life. This reminds us of certain biblical precepts relative to conjugal life, and the too neglected theory of M. Thury (of Geneva) on the influence of the degree of ovular maturity on the production of the sexes. But spontaneous oscillations in the proportion of the sexes are always feeble; even the matrimonial type does not seem to influence them, for in the harems of Siam the sexual relation of births is the same as in Europe.[171] On the other hand, it is proved that race-horses, which are very polygamous, since they serve as stallions, have male and female descendants in exactly equal proportions.[172]