and there is consequently nothing to represent the equivalent of a geological section. Such, I trust, is a fair rendering of the gist of Dr. River's contention.
Now there are probably more ways than one in which this contention might be met, but the simplest of all would seem to be the following. One has only to point out that the procedure of a science must adapt itself to the nature of its special subject-matter. One sort of subject-matter invites one line of treatment, and another another. The logic must be accommodated to the facts, not the facts to the logic. Thus the crust of the earth lends itself readily to the stratigraphical method of exploration. Geological considerations of this nature must have impressed themselves on the mind of Adam when first he struck a spade into the ground. On the other hand, as we have just seen, the savage who is destitute of a recorded history does not lend himself to be viewed in section by the casual observer. A spade makes no impression on the hard soil, and operations are perforce confined to the surface. It is as if a geologist had to depend for his determination of the invisible strata below ground on the uncertain indications afforded by what he takes to be their several outcrops,—a method which would at once oblige him to call in the aid of those physical theories which, according to Dr. Rivers, should normally serve him but as a sort of savoury after meat. Survivals of culture are just such outcrops showing at the surface. They are at first sight as much a part of the existing life of the people under investigation as is the rest of their customs. To the discerning eye of the theorist, however, they admit of correlation with otherwise hidden strata,—with former land-surfaces, as it were, that are now for the most part obliterated beneath the accumulations of tinie,—because of their present uselessness. But what is the test of such uselessness? Is it the observer's sense of fitness and convenience, or is it that of the people concerned? But if you decide, as surely you must, that the
