and the annual Saturnalia. (W.  Crooke, “The Hill Tribes,” Journal of the Anthropological Institute, new series, vol. i, 1899, p. 243.) The association of dancing with seasonal ritual festivals of a generative character—­of which the above is a fairly typical instance—­leads us to another aspect of these phenomena on which I have elsewhere touched in these Studies (vol. i) when discussing the “Phenomena of Periodicity.”

The Tahitians, when first discovered by Europeans, appear to have been highly civilized on the sexual side and very licentious.  Yet even at Tahiti, when visited by Cook, the strict primitive relationship between dancing and courtship still remained traceable.  Cook found “a dance called Timorodee, which is performed by young girls, whenever eight or ten of them can be collected together, consisting of motions and gestures beyond imagination wanton, in the practice of which they are brought up from their earliest childhood, accompanied by words which, if it were possible, would more explicitly convey the same ideas.  But the practice which is allowed to the virgin is prohibited to the woman from the moment that she has put these hopeful lessons in practice and realized the symbols of the dance.”  He added, however, that among the specially privileged class of the Areoi these limitations were not observed, for he had heard that this dance was sometimes performed by them as a preliminary to sexual intercourse. (Hawkesworth, An Account of the Voyages, etc., 1775, vol. ii, p. 54.)

Among the Marquesans at the marriage of a woman, even of high rank, she lies with her head at the bridegroom’s knees and all the male guests come in single file, singing and dancing—­those of lower class first and the great chiefs last—­and have connection with the woman.  There are often a very large number of guests and the bride is sometimes so exhausted at the end that she has to spend several days in bed. (Tautain, “Etude sur le Mariage chez les Polynesiens,” L’Anthropologie, November-December, 1895, p. 642.) The interesting point for us here is that singing and dancing are still regarded as a preliminary to a sexual act.  It has been noted that in sexual matters the Polynesians, when first discovered by Europeans, had largely gone beyond the primitive stage, and that this applies also to some of their dances.  Thus the hula-hula dance, while primitive in origin, may probably be compared more to a civilized than to a primitive dance, since it has become divorced from real life.  In the same way, while the sexual pantomime dance of the Azimba girls of central Africa has a direct and recognized relationship to the demands of real life, the somewhat allied danses du ventre of the Hamitic peoples of northern Africa are merely an amusement, a play more or less based on the sexual instinct.  At the same time it is important to bear in mind that there is no rigid distinction between dances that