Page:Folk-lore - A Quarterly Review. Volume 20, 1909.djvu/295

Rh of other differences which characterise what may be called the Nanga region, and yet such differences undoubtedly exist. Some of the mountain tribes which live in or near this region possess a kinship system of an order wholly different from that of the coastal tribes of which Mr. Thomson gives a full account. Further, the institution of cross-cousin marriage, usually regarded as typically Fijian, is absent in these tribes, and the form of totemism is also different from that of the coast. Such differences are vital; they do not concern the superficialities which pass so easily from tribe to tribe, but they are at the very basis of the social and religious life of the people, and point to the widest differences in origin or environment, and yet from Mr. Thomson we do not even gain a hint of their existence.

It must be acknowledged that in his preface the author disclaims any attempt to give an exhaustive account of the Fijians, and an apology should perhaps be offered for having made this book a means of drawing attention to a frequent defect of anthropological literature. Though, however, the author may only have intended his book to be a study of decay of custom, he has as a matter of fact given so good an account of the people that a word of warning is necessary to indicate that Fiji has not yet made its full contribution to our knowledge. A detailed study of this most important meeting place of the Melanesian and Polynesian cultures will yet reveal much of the utmost importance to the student of human society.

a paper published under the above title, Mr. Lang returns to his favourite subject, the social organisation of the Australian tribes. Remarking that the classificatory system of naming relations is still, perhaps, the strongest card in the hand of believers in a period of human promiscuity, he discusses how far the