forth to amuse a circle of listeners." That some myths may have arisen from words, the perennial tendency to indulge in folk-etymologies may well induce us to believe: the question is, how many, and which? The danger of speculations of this kind is that they are apt to be mere guesses, of no more value than folk-etymologies themselves. Nor does Professor Brinton's division of stories, as here given, exhaust the distinctions. Beside sacred sagas and märchen (including drolls) there are what we may term secular sagas—stories told in good faith as true, but in their origin neither cosmical speculations nor relating to the doings of divine personages—many of which may ultimately become sacred. All three classes of tales are probably derived from a single kind of narrative believed to be true, from narratives in which there was no dividing line between natural and supernatural, between human and extra-human. The savage may now recognise the difference between sacred sagas and märchen; but the want of differentiation even yet between the religious and other aspects of savage life points to a difficulty in the theory of the original distinction between stories with a directly religious intent and other narratives, which we think the learned author has not fully considered.
Dr. Brinton's criticism of the terms fetish and fetishism follows similar lines to that of Dr. Jevons; and it were much to be wished that the result might be to drive these terms from scientific writing, or at least to confine them to savage amulets and the use of savage amulets, particularly Negro and Bantu. The observation on totemic animals or eponymous ancestors, as "not to be taken literally," is more doubtful. "They were not understood," we are told, "as animals of the sort we see to-day, but as mythical, ancient beings, of supernatural attributes, who clothed themselves in those forms for their own purposes." It is impossible to read either savage tales or descriptions of savage rites without being led to the conclusion that impermanence of form is a cardinal belief in the lower culture. Not merely does the savage find it impossible to draw a line between the brute creation and humanity, but he seems to go further, and to hold that the same personage, now equally as in former ages, may appear sometimes and in some circumstances as a human being, and at other times and in other circumstances as a wolf, a snake, a tree, or even a stone. Identity of personality seems to be quite independent of shape. Hereto we may trace the survivals in comparatively civilised society of
