The rise of the goddesses may have been due in part to the influence of Dravidian folk-religion. This does not, however, vitiate the theory that moon, water, and earth worship was not unconnected with the ascendancy of the Brown race in India. The Dravidian brunet long heads were, as we have said, probably represented in the pre-Aryan, as well as the post-Vedic folk-waves, which mingled with pre-Dravidian stocks. Mr. Crooke inclines to the view that the Aryan conquest was more moral and intellectual than racial.36 The decline of the patriarchal religion of the Vedic military aristocracy may thus be accounted for; the religious practices of the earlier people might ultimately have attained prominence in fusion with imported ideas. If the Aryan racial type was distinctive, as it appears to have been, in colour at any rate, the predominant people who flourished when the hymns were composed, may have greatly declined in numbers owing to the ravages of disease which in every new country eliminates the unfit in the process of time. Even if Aryan conquest was more racial in character than Mr. Crooke will allow, the physical phenomena of the present day can be accounted for in this way, due allowance being made, of course, for the crossment of types. In all countries which have sustained the shock of invasion, the tendency to revert to the aboriginal type is very marked. At any rate, this is the case in Egypt and Crete as present-day evidence shows. In Great Britain, which was invaded by the broad heads of the Bronze Age, the long-headed type is once again in the majority; a not inconsiderable proportion of our people show Stone Age (Mediterranean) physical characteristics.

      In this connection it is of interest to refer to immemorial beliefs and customs which survive in representative districts in Britain and India where what may be called pre-Aryan influences are most pronounced. A people may change their weapons and their language time and again, and yet retain ancient modes of thought. In Devon, which the philologists claim to be largely Celtic like Cornwall, the folk-lore shows marked affinities with that of Ireland, Wales, and Scotland, suggesting the survival of ancient Mediterranean racial influence, for much of what we call Celtic links with what belongs to ancient Greece and the Egyptian Delta. Mr. Gomme has shown37 in an interesting summary of recorded folk-practices that the “ram feast” of Devon resembles closely in essential details similar ceremonies in ancient Greece and modern India. At the beginning of May the people of Devon were wont to sacrifice a ram lamb to the deity of waters. The animal was tied to a pillar, its throat was cut, and young men scrambled to obtain pieces of its flesh for girls. The devourer was assured of good luck during the year. After the ceremony, dancing, wrestling, and drinking were indulged in. A comparison is drawn between this and similar rites among the ancient Semites and ancient Greeks. In India a Dravidian Paria acts as the temporary village priest. He uses a whip like the “gad whip” in Lincolnshire, and kills the lamb by tearing its throat with his teeth. A scramble takes place for the flesh, the people circulate the village, as some communities in our own country still perpetuate the ceremony of “riding the marches” of ancient burghs; then universal licence prevails. Similarly law was suspended at the ancient Scottish Hallowe'en celebrations; in some districts even in our own day Hallowe'en and New Year practical jokes and rowdyism is still prevalent. Herodotus refers to the universal licence and debauchery which characterized the Isis festival in Egypt.

      A remarkable feature of post-Vedic religion in ancient India is the prominence given to the doctrine of metempsychosis (transmigration of souls) and the conception of the yugas or ages of the universe.

      In the Rigveda the soul of the dead proceeds at once, or at any rate after burial, towards the next world. In one passage only is it spoken of “as departing to the waters or the plants”, and this reference, Professor Macdonell suggests,38 “may contain the germs of the theory” of transmigration. In the speculative prose treatises, the Upanishads, which were composed in the Middle Country, the doctrine of metempsychosis is fully expounded. It does not follow, however, that it originated in India although it may have obtained there unrecognized by the priestly poets who composed the hymns to the deities, long before it became an essential tenet of orthodox or official religion. Other representative communities of the Brown race professed this doctrine which appears to have evolved from the vague belief shared by more than one primitive race, that the souls of the dead, and especially of dead children, were ever on the outlook for suitable mothers. Even in Central Australia a particular tribe has perpetuated “the germs of the theory”, which may also be traced in the widespread custom of visiting standing stones at a certain phase of the moon to perform a ceremony so that offspring may be obtained. The Upanishadic doctrine of metempsychosis is less likely to have been so much coincidental as racial when we find that it is restricted to those areas where definite racial influences must have been at work. The Greeks believed in transmigration. So did also a section of the Egyptian people as Herodotus has stated and as is proved by references in folk-tales, temple chants and inscriptions.39 As we show (Chapter VI), the Irish conception closely resembled the Indian, and it also obtained among the Gauls. There is no trace, however, that the Teutonic peoples were acquainted with the fully developed doctrine of metempsychosis; the souls of the dead departed immediately to Valhal, Hela, or the loathsome Nifelhel.

      The doctrine of the world's ages is common to the Indian, Greek, and Irish mythologies, but is not found in Teutonic mythology either.40 There are indications that it may have at one time obtained in Egypt, for there was an Age of Ra, then a deluge, an Age of Osiris, an Age of Set, &c.; but the doctrine, like other conceptions in Egypt, probably suffered from the process of priestly transformation in the interests of sectarian propaganda.

      In India the ages are called the yugas, and this term has a totally different meaning in Vedic and Upanishadic times. Evidently the Bharata invasion and the establishment of the middle country power of their allies, the Kuru-Panchalas, was not unconnected with the introduction of the doctrines of metempsychosis and the yugas, and the prominence subsequently given to the worship of female deities.

      If this theory can be established, we are confronted by an extremely interesting problem. It would appear that the mythology of the Vedic period bears a close resemblance to Teutonic, while that of the post-Vedic period connects more intimately with Greek, Celtic, and Egyptian. Assuming that the Vedic people were influenced by what we recognize as Teutonic modes of thought, do we find here proof that the Aryans came from Europe? In Chapter II it is shown that the Norse Heimdal displays points of resemblance to Agni. The former, however, has been developed almost beyond recognition as a fire god, and it is evident that we find him in northern Europe in his latest and most picturesque form. On the other hand, there is no dubiety about the origin of the Vedic Agni.

      The evidence afforded by archæology is highly suggestive in this connection. Scandinavia received its culture from the south at a comparatively late period in the Bronze Age, and it