These conclusions are most fully corroborated by evidence extraneous to this paper, which further amplifies our knowledge by showing that the invasion of the Au-rai-ria folk came from the West, and affected at least eleven of the sixteen Gilbert Islands.
In the Nui and Banaban creation tales we therefore seem to have an illustration of the process by which history declines into myth. It is only because we are in possession of material for cross-checking, and of extraneous knowledge into the bargain, that we are able to recognise it as history at all. Had we not come by such knowledge, we should unquestionably have called it pure myth, and classed it, say, with the tale of the dethronement of Cronus, or any other conflict of gods recorded in primitive lore. If, then, we may dare, on so small a collection of facts, to draw any inference at all, it is that what passes under the name of myth might very often prove to be decayed history, if all facts were known. Or, putting this generalisation in another form, it would seem that the mere record of fact, apart from its exegesis, has been instrumental in the elaboration of race mythologies, as we view them to-day.
II.
Before passing to other matters, I should like to answer several questions which arise from the above discussion of race migrations into the Gilbert Islands. This will bring us back to the text of the creation-myth which I have exhibited. That version propounds the dogma that Samoa was the first land created, the home of the ancestral Tree, and the pristine soil of the Gilbertese forefathers. Enough of its historic appendices has been outlined to show that the case is upheld with a wealth of genealogical detail. Its contention is, in fact, up to a certain point correct. There can be no doubt whatever, that the last victorious invasion of the Gilbert Islands came from Samoa; the modern race
