three parts of the festival was forgotten, and the spectators were reduced to their own conjectures. The leading fact and the starting-point for all attempts at explanation was that the festival was in honour of the god Dionysus, and whatever was done or represented in it must be something redounding to his glory. Then who was represented by the figure on the tree-top which was treated with such hostility and hatred, pelted and pulled to pieces by the women? It must be some enemy of the god, whose destruction was a triumph for Dionysus and was therefore commemorated in this festival. The women evidently were on the side of the god-must have been his worshippers-therefore the man was not one of Dionysus' worshippers. Perhaps that accounts for the opposition between him and the god: he would not bow down to Dionysus, whereas the women accepted the god cheerfully-the women of a community would be more likely to welcome a novelty in worship than the head of the family and representative of the old worship. But why is the man dressed in woman's clothes? no man in his senses would go about in public dressed up like a woman. No; but it is just one of the powers of the wine-god that he makes men lose their senses-and that may account, too, for the women killing their own king, they must have been frenzied to do that. So there only remain two things not clear now, why is the god not represented at his own festival? and what is the meaning of the tree being suddenly hauled up erect? Perhaps the god is supposed to be present, invisible but directing everything; and in that case it is he who causes the tree-top to rise, after inducing his foe to mount it, in order that, after exposing him to ridicule, he may cause him to perish at the hands of the women of his own family.

We have only now to fill in the proper names in order to have the myth of Pentheus which affords the framework of Euripides' play, the Baccha. Pentheus is the king who resists the introduction of the worship of Dionysus,1 and is consequently bereft of his senses and led in woman's clothes as a laughing-stock through his own town by Dionysus.

1I explained the similar myths of Lycurgus, Eleuthera, and Tiryns in much the same way in Folk-Lore, June 1891, vol. ii. ii. 238-41.

The women of Thebes, headed by Agave, the mother of Pentheus, are the women who accept the god, and become mænads. It is to enable Pentheus to spy their worship that Dionysus bends down a pine-tree, sets him on the top, and then lets it go. Finally, it is Agave who, with the other bacchæ, pelts and pulls to pieces her own son and carries off his head and sets it on his own palace-gable.

The tendency of syncretism to yield myths is not confined to Greece. Let us take a pair of instances from the New World. The Chibchas of New Granada had a goddess who dwelt as a serpent in Lake Iguaque, but whose name, Bachuê, "simply means 'she who suckles the maize,'" 1 i.e. she was a maize-mother, a plant totem, from whom the Chibchas traced their descent. Evidently the worshippers of this maize-mother had united their worship with that of a clan having an animal, a serpent, for totem; and the worship of the water-spirit had further been incorporated with that of Bachuê, with the result that a myth had to be invented to account for it all, and was to the effect that on the first day of the world there emerged from its [Lake Iguaque's] waters a beautiful woman named Bachuê or Fuzachagua [= the good woman], carrying in her arms a child three years old. These were the ancestors of the race: when the world was peopled, they returned to the lake, and disappeared in its waters in the form of serpents."2 The syncretism of a maize-goddess and a bird-totem has given rise to the myth told by the Cañari Indians, in the district southward of Quito. There were once two brothers whose provisions were exhausted; "the herbs and roots which they were able to collect scarcely sufficed for their sustenance, and hunger sorely pressed them, until two parrots entered their hut in their absence and prepared them a meal of cooked maize, together with a supply of the fermented liquor (chicha), which is made by steeping it in water. This happened day by day, until at length one of the birds was made captive by the brothers. When thus captured, it changed into a beautiful woman, from whom the brothers obtained the maize-seed and learned the art of cultivating it, and who ultimately became the ancestress of the Cañari nation."3 Possibly the maize 1 Payne, New World, i. 455.

1

Ibid.

3 Ibid. 327.

was originally the totem of the women, the parrot of the men, of the tribe; for the cultivation of maize, Mr. Payne adds, “ was in the earliest times the exclusive task of the women of the tribe. It is only in a later stage that it is shared by the men," and then the men would be admitted to the worship of the maize-goddess, and the maize totem would be placed by the side of the parrot totem, till the worship of the two blended in one whole, and required a myth to explain it.

There was a time in the history of man when as yet the first tale had not been told, and the very idea of story-telling had not yet occurred to his mind. When it did occur, it was probably due to suggestion and not to his own unaided invention; and probably also it was an idea of very slow and gradual growth. The explanations which primitive man found for the various problems which perplexed him were of course, to him, actual facts, not pieces of imagination; and they were mostly single incidents, usually destitute of interest except for the community for whom they were originally designed-they might and did supply materials for tales, but they were not themselves tales. Some of these explanations, however, being designed to explain a series of phenomena, would spontaneously form a series of incidents, forming a true tale, e.g. as in the case of the myth of Pentheus; and some, as for instance the Cañari myth, would have a charm of their own which would win and delight other people besides the actual descendants of the bird-maiden. The man whose memory affectionately retained as many of these myths as he could gather, and who could repeat them well, would always command an audience. When he had told all he could easily remember, the tribute of praise couched in the appealing imperative, "Go on!" would stimulate him to rack his memory, with the result that semi-consciously he might substitute for the original incident or character some analogous one-the transformation into an animal instead of a bird, a god for a goddess, a jealous Hera for an irate Dionysus-and when what was first done semi-consciously came to be done with full consciousness and deliberation, the art of story-telling would be accomplished. Again, tales with a permanent human interest would easily spread beyond the limits of the original audience, and so would tend to

become detached from the belief or ritual or other institution which they were first invented to explain. But in such circumstances statements which were in the first place explanations of something come themselves to require explanation: the Kalang chief was transformed into a dog, or a maiden into a bird, but why? The question was inevitable, and the answer would add a fresh incident to the story, a fresh complication to the plot. Further, the answer would be sought amongst incidents already familiar to the narrator and his audience, or would be framed on the analogy of one of them. Now, of such incidents there would be plenty that had been framed by early man to account for the numerous problems which interested him. One such problem was raised by taboos: to approach certain persons under certain circumstances, mourners, women, and others, was tabooed, but why? because once someone violated the taboo, and he or the tabooed person suffered a certain dreadful thing-in folk-tales the tabooed wife is often changed into a serpent or a bird.1 Now, deities who confer benefits on man, teach him to cultivate maize for instance, frequently disappear, when they have completed their beneficent work—sometimes, like Bachuê, disappear in animal form. Here we have a series of very easy "chances" for the story-teller ransacking his memory: the parrot-maiden who married a human being. eventually departed as she came in the shape of a bird, and so departed because her husband violated a certain taboo. Such a story would be interesting even to those who did not claim to be descended from the heroine, and were not interested in the cultivation of maize. It would be interesting enough to spread, vivu' volitare per ora virum. And as a matter of fact it is the type of a class of tales found all over the world, and known as Swan-maiden tales, from the bestknown example, the Arabian Nights' tale of "Hasan of Bassorah." 2

The incidents which compose the Swan-maiden story are such as have been familiar probably to every race at a certain stage of its development, and accordingly-unless we make the somewhat arbitrary and certainly unproved assumption 1 Lang, Custom and Myth,2 75 ff.

2 See Hartland, Science of Fairy Tales, cc. x. and xi.

that there was only one race of men capable of telling tales -those incidents may have been put together in this particular combination at any place in the inhabited world. But it does not follow that this particular combination would be formed by every race which was familiar with the separate incidents. The Cañari combination might indeed spring up independently in several centres, for a number of tribes trace their descent to the maize-mother or a cereal goddess, and the circumstances which would lead to a belief in the transformation of the goddess into an animal are fairly common also; and the particular animal might be a bird in several cases; or, if it was a serpent, then we should get a tale of the Mélusine class. But the further incidents of the departure of the beneficent deity, and in animal form, and that form a bird (or serpent), and that departure in consequence of a violation of taboo-though they might conceivably have been combined in this particular sequence more than once, probably are not, on the theory of chances, likely to have come together in this particular form. When, then, we find the story with its full complement of incidents (or in a form which clearly postulates the previous existence of the full complement) in several different places, we should conclude that it has spread to them from its place of origin. We have, then, now to consider the problem of the diffusion of myths.

One way in which a myth might be diffused is the dispersion of the people to whom it was known. The IndoEuropeans spread from their original home, wherever that was, until they covered Europe and part of Asia; and if they had any tales interesting enough to live, those tales may well have been diffused over all the area eventually covered by the Indo-Europeans. But it is quite certain that the circulation of those tales would not be confined to the Indo-European public: they would find their way to all peoples with whom the Indo-Europeans had dealings, and there would be an international exchange of tales as well as of goods. In other words, borrowing is a factor in the diffusion of myths as well as tradition. And when we reflect that the Oceanic or Malay race has come to extend from the Sandwich Islands on the west to Madagascar on the east, and