Page:Folk-lore - A Quarterly Review Volumes 32 and 33.djvu/412

104 already pricked out by Percy Smith, but also down the chain of central Pacific archipelagoes, in the manner conjectured by Churchill. They would thus seem to have followed the route of the Proto-Polynesians (Ur-polynesier) postulated by Thilenius.

This is the merest sketch of a thesis, of which the adequate presentation demands a whole chapter, and the argument an entire volume; I have only ventured to give the subject a place in this paper in order to indicate the possibilities that exist in the comparative study of Gilbertese traditions. I do not claim that from the oral records of any islanders a complete history of their racial affinities may be traced: the evidence of tradition, especially in these days of decay, is necessarily discrete. But I do urge that since the concrete method of sociological research has been proved so efficient a tool by Dr. Rivers, the value of tradition as a clue to race-fusion and race-migration has been ignored, and perhaps deliberately discounted, by students of anthropology; I would therefore lay stress upon the fact that many inferences as to race-movements in Oceania, drawn exclusively from the study of Polynesian tradition by such writers as Fornander and Percy Smith, have been amply corroborated by sociological research. Certainly such inferences have been also, in the main, strengthened and clarified by the new evidence; the modern method is an enormous scientific advance on any of its predecessors, but it cannot afford to exclude the study of tradition from its schedule of useful subjects.

The various myths, in which the cosmogonic ideas of the Oceanic peoples have been preserved, are separated by