in this splendidly exhaustive treatise. The way is therefore prepared for marriage to be treated as part and parcel of a larger group of institutions. In so far as it is founded upon natural instincts in man its features may be traced through all human societies; in so far as its forms have been affected by social requirements its features must differ according to the grades of social development with which it is associated.
Mr. Westermarck lays almost too much stress upon some of the natural features of marriage, at all events in so far as they are used as materials for its history. For instance, man as the nourisher of his wife and offspring is considered at some length, and evidence is produced from a great number of savage and barbarous peoples, ranging from the Fuegians up to the Arabs. When, therefore, we meet in folk-lore such a custom as Miss Burne mentions as obtaining in Shropshire—"if a husband failed to maintain his wife she might give him back the wedding-ring, and then she would be free to marry again" (p. 295)—how are we to arrange and classify this survival? The effect of such a practice would lead us back to a state of temporary monandry, and would not account for the beginning of permanency in the marriage-tie. If the condition of man as the nourisher is put forward as a vera causa for the hypothesis that in primitive times man, woman, and children, formed a recognizable social unit, the supporters of such a hypothesis must answer the obvious objection suggested by the piece of Shropshire folk-lore, that when he ceased, either from inability or caprice, to nourish, the social unit of which he was a necessary element went to pieces.
On the other hand, some of the forms resulting from the effects of a conscious use of natural marriage for social organisation are scarcely treated with sufficient length. Thus, the bars to marriage between members of different races are set forth in some detail, and the evidence is most important; but the corresponding evidence of marriage
