(M171) We have now concluded our inquiry into the nature and worship of the three Oriental deities Adonis, Attis, and Osiris. The substantial similarity of their mythical character justifies us in treating of them together. All three apparently embodied the powers of fertility in general and of vegetation in particular. All three were believed to have died and risen again from the dead; and the divine death and resurrection of all three were dramatically represented at annual festivals, which their worshippers celebrated with alternate transports of sorrow and joy, of weeping and exultation. The natural phenomena thus mythically conceived and mythically represented were the great changes of the seasons, especially the most striking and impressive of all, the decay and revival of vegetation; and the intention of the sacred dramas was to refresh and strengthen, by sympathetic magic, the failing energies of nature, in order that the trees should bear fruit, that the corn should ripen, that men and animals should reproduce their kinds. (M172) But the three gods did not stand by themselves. The mythical personification of nature, of which all three were in at least one aspect the products, required that each of them should be coupled with a goddess, and in each case it appears that originally the goddess was a more powerful and important personage than the god. At all events it is always the god rather than the goddess who comes to a sad end, and whose death is annually mourned. Thus, whereas Osiris was slain by Typhon, his divine spouse Isis survived and brought him to life again. This feature of the myth seems to indicate that in the beginning Isis was, what Astarte and Cybele always continued to be, the stronger divinity of the pair. Now the superiority thus assigned to the goddess over the god is most naturally explained as the result of a social system in which maternity counted for more than paternity, descent being traced and property handed down through women rather than through men. At all events this explanation cannot be deemed intrinsically improbable if we can show that the supposed cause has produced the very same effect among existing peoples, about whose institutions we possess accurate information. This I will now endeavour to do. § 2. Influence of Mother-Kin on Religion.