belief in them as an excrescence on religion in a secondary stage of its growth, to which it would be easy to attach undue importance. I would not be understood, however, to contest the doctrine that the notion of a more or less spiritual existence of the dead has been a cardinal factor in early religious development. No reader of Primitive Culture can doubt this. The case of Japan raises a presumption that a considerable advance must be made towards a Spiritist form of faith before ghosts can make their appearance. The modern popular literature, written after centuries of Buddhist and Chinese influence, teems with apparitions.
The deification of human beings, by which something of the superhuman power and glory already recognised in natural deities is reflected back upon heroes, ancestors, or sovereigns, does not occupy an important position in Shinto. As already pointed out, the ancestral gods are not really deified ancestors but existing deities who have been converted into ancestors, or others invented for this very purpose. The deification of living and deceased mikados and princes belongs to a comparatively recent period, and is open to strong suspicion of Chinese influence.
There is no summer and winter myth in these old records, no rainbow myth, and no eclipse myth. There is, strange to say, no earthquake myth, and but one solitary mention of a god of earthquakes. The most terrible exceptional convulsions would appear to have impressed the religious sense of the Japanese less than every-day, normal phenomena.
