We have to choose between the two explanations given above. I am, on the whole, disposed to agree with Dr. Mannhardt, and in the absence of convincing evidence as to the regular and periodical occurrence of human sacrifice in ancient Italy, to regard these strange survivals as semi-dramatic performances rather than sacrificial rites. This view, however, need not exclude the possibility of the union of both drama and sacrifice at a very remote period, probably before the Latins settled in the district. The immersion in water, whether or no it involved the death of a victim, is reasonably explained, on the basis of comparative evidence, to have been a _rain-spell_[465]. In the cases already mentioned of Adonis, Nerthus, &c., this idea seems the prominent one. I am inclined to think, however, that the notion of purification was also present—the two uniting in the idea of regeneration. Plutarch calls the Argean rite ‘the greatest of the purifications,’ and he is here most probably reproducing the opinion of Varro[466]. This is indicated by the presence of the priests and the Vestals, by the processions, and by the mourning of the Flaminica Dialis, as we have already seen. We may regard the rite as in fact a casting out of old things, and in that sense a purification; and also at the same time as a spell or earnest of rain and fertility in the ensuing year. The puppets were perhaps hung in the sacella in the course of the procession in March, as a symbol of the fertility then beginning, and cast into the river as ‘the old ones’ when that fertility had reached its height[467]. In the last place, it might be asked in honour of what deity the rite was performed. It is hardly necessary, and certainly is not possible, to answer a question about which the Romans themselves were not agreed. Ovid and Dionysius[468] believed it was Saturnus, probably following an old Greek oracle which was known to Varro[469]. Verrius Flaccus thought it was Dis Pater[470]. Modern writers have concluded on the general evidence of the rite that it was the river-god Tiberinus; Jordan, however, regarded the question as irrelevant[471]. We may agree with him, and at least return a verdict of _non liquet_. If it was a sacrificial act, the ancient river-god is indeed likely enough; if it was a quasi-dramatic one, it does not follow that any deity was specially concerned in it. But we may go so far as to guess that it was connected with the worship of those vaguely-conceived deities of vegetation whose influence on the calendar we have been tracing since March 1.