Since the first publication of Custom and Myth, many other works have appeared, dealing on the same principles with matters of belief, fable and ritual. Were the book to be re-written, numerous fresh pieces of evidence might be adduced in support of its conclusions. In Mr. Frazer’s Golden Bough (Macmillan) the student will find a carefully conceived argument, and a large collection of testimonies, bearing on the wide diffusion, among savages and civilised peoples, of ancient rites and ancient ideas. The works of Mannhardt have practically been introduced to the English reader by Mr. Frazer, with much new matter of his own. The main topics are the worship of human gods and the superstitions connected with vegetation. To push a theory too far is the common temptation of mythologists, and perhaps Mr. Frazer’s cornstalk does rather threaten to overshadow the whole earth and exclude the light of sun and sky. But the reader, whatever his opinions, will find great pleasure and profit in Mr. Frazer’s remarkable studies, and in those of Mannhardt, which were unknown to myself when I wrote Custom and Myth.

In Miss Harrison’s volume on Athenian Myths the student will find the ?tiological theory (namely, that many myths were invented to explain obscure points of ritual) applied in a number of classical instances. A singularly ingenious study of Roman myths is presented in Mr. Jevons’s edition of Plutarch’s Romaine Questions (Nutt). These are recent instances of the use of the ‘anthropological’ method, first firmly established by Mr. Tylor’s Primitive Culture, and now holding its own as a recognised instrument in the study of the historical development of the imagination. In Rosscher’s Ausführliches Lexikon of Greek and Roman mythology, the earlier method of the philologists is usually adopted, and the work, still in course of publication, is most useful for its recondite learning.

These notes are meant for the guidance of any reader who may care to push his studies further than the sketches of the present volume.

On one or two points some remarks may be necessary. The author has been not unnaturally accused of seeing Totems everywhere. He would therefore protest that he does not regard every beast and bird which appears in myths or in religious art as necessarily a Totem. But he inclines to think that where Celts or Greeks claim descent from a god who pursued his amours in animal shape, or where a tribe bears the name of an animal, regards that animal with religious respect, and places its effigy beside that of a god, the Totemistic hypothesis colligates the phenomena, and deserves consideration. These and other early features of religion occur mainly in Greece after the Homeric age. It has been suggested, for example, by Mr. Walter Leaf, that Homer’s people, the Ach?ans, were free from all such ideas as Totemism, worship of the dead, ritual of purification for homicide, the mysteries, and so forth. These were notions held by the Pelasgi, and revived or retained by the Ionians, an older and distinct stock of Pelasgian origin. I am unable to convince myself in this matter, not knowing how much of the refinement in the Homeric poems is due to the genius of the poet, who might ignore practices with which he was familiar. They may have been Pelasgo-Ionians, who derived Helen’s birth from the Swan, or Homer may have chosen to slur over an Ach?an legend, and so on in other cases; for example, as to the descent of the Myrmidons from Zeus in the shape of an Ant. On another point a word may be said. One has been accused of believing that identical popular tales, the same incident in the same sequence of plot, might arise simultaneously in savage imaginations in all parts of the world. In Custom and Myth it will be plain that I say nothing of the sort. ‘The Far-Travelled Tale’ is one instance chosen to show that such a story must probably have drifted, somehow, round the world. On the other hand, in ‘Cupid and Psyche,’ it is asserted that the central incident might be invented wherever the nuptial taboo on which it is based was recognised. The exact sequence of incidents in the ‘Cupid and Psyche’ of Apuleius, on the other hand, could probably only be invented once for all. But we find the central incident where we do not find the sequence of incidents which make up ‘Cupid and Psyche.’ A full statement of my ideas is prefixed to Miss Roalfe Cox’s Cinderella (Folklore Society). As a rule, the incidents in M?rchen are common to all races; an artistic combination of many of these in a plot must probably be due to a single imagination, and the plot must have been diffused in the ways described in Custom and Myth. Independently evolved myths may closely resemble each other when they account for some natural phenomenon, or are based on some common custom. Wherever a sequence of such incidents is found in a distinct and artistic plot, we may provisionally assign diffusion from an original centre as that cause. Singular as are the coincidences of fancy, it is unlikely that they ever produced exactly the same tale in lands which have never been in communication with each other. I am unable to conjecture why Mr. Jacobs, M. Cosquin, and probably other critics, regard me as maintaining that all similar tales in all countries have been independently evolved. I have always allowed for the possibility both of diffusion and, to a certain extent, of coincidence, as in the Red Indian forms of ‘Cupid and Psyche’ and of ‘The Dead Bride,’ a shape of the story of Eurydice. Discussion would be simpler, if controversialists took the trouble to understand each other.

In the Report of the Folklore Congress of 1891 (p. 65) I find that I said ‘the suggestion that exactly the same plot, in exactly the same shape, and with exactly the same incidents, can have been invented by several persons independently, seems to me inconceivable,’ and on p. 74 I find M. Cosquin alleging that my opinion is the very reverse, followed by Mr. Jacobs (p. 85). I have tried to explain that I believe in no such exact coincidences of imagination, though how far precisely coincidence may go is a delicate question.