Page:Folk-lore - A Quarterly Review. Volume 15, 1904.djvu/382

 354 Reviews.

stantially accurate. A curious feature is that they go back in most cases four or five generations and then stop abruptly. It is im- possible, in the face of the accuracy of the genealogies actually recorded, which reach back at least a hundred years, to regard this as due to a failure of memory. Dr. Rivers makes the very probable suggestion that it has something to do with the introduction of patrilineal descent.

The third section, also by Ur. Rivers^ deals with the system of kinship, and is based on the genealogies. It is impossible within brief limits to discuss the system, which is of course classificatory, or to consider the ingenious suggestions made by Dr. Rivers as to the mode in which the meaning of various terms was extended. In connection with kinship terms a careful analysis of name tabus and avoidance is given. For a man the names of parents-in-law were of course tabu ; so, too, were the names of the wife's uncles and aunts, real and by marriage, of her grandparents, and great uncles and aunts and their children, and of the paternal nephew's wife, all of whom bear the kinship name ira ; the sister's husband, cousin's husband, wife's brother, wife's cousin's husband, and wife's male cousin were imi, and their names were tabu ; like the first group they sometimes conversed in a low tone ; finally the brother's wife, cousin's wife, wife's sister, wife's cousin's wife, and wife's female cousin were ngattbat, and this relationship too was a bar to conversation to some extent.

The avoidance customs were comparatively indefinite, and, as in the case of some Australian tribes, the mother-in-law was not hedged round with more severe rules than other relatives.

A highly interesting group of customs, which are, however, not dealt with fully, are those connected with the functions of certain kin. The maternal uncle in particular has special rights over his nephew, and has the power of stopping a fight in which he is engaged. With this may be compared the New Caledonian cus- toms by which the maternal uncle acquires extensive rights over the property of a nephew whose blood is shed in his sight {Les Miss. Cath., 1880, p. 68).

The fourth section, by Dr. Haddon and Dr. Rivers, deals with totemism. A careful study of the local distribution of the clans has shown that segmentation has almost certainly taken place. Various interesting points are brought out in connection with the rise of the Sam (cassowary) clan, which seems to be intrusive, and