The religious customs of those peoples who were relatively aboriginal to the Nahua support the theory of the predominance of the rain-cult in Mexico from a very early period, and Torquemada states that during seasons of drought the Otomi sought to propitiate the rain-gods by sacrificing a virgin on the top of a hill. [29] Espinosa says that the Tarascans sacrificed snakes rather than human beings—possibly for the same reason as the Esquimaux beat their dogs during an eclipse, in order that the Great Dog which causes the undesirable phenomenon may desist, the Tarascans probably killing the reptiles in question in order that the Great Snake might relent and send rain. [30] The towns about Chapala paid divine honours to the spirit of the adjacent lake. Late though these survivals may have been at the era of the Conquest, yet they seem to have enshrined the memory of an early rain-cult among the peoples with whom they were found, and many others could be adduced. THE QUETZALCOATL CULT The appearance of the Quetzalcoatl cult in Mexico, which would seem to have entered the country at some time about the middle of the eighth century, must have caused very considerable alterations in the simple and probably as yet uninfluenced rain religion which it found in occupancy. From whatever portion of the Isthmian tract it came, one thing regarding it is positively certain—that it introduced into Mexico the rudiments of the calendric computation evolved in Central America. In its phase as imported by the apostles of the Quetzalcoatl religion, it seems fairly certain that the tonalamatl was of the nature of a lunar time-count, and the probabilities are that its protagonists discovered on their arrival in Anahuac that a count similar in character was in use among the priesthood of the Tlaloc worshippers, who as an agricultural people could hardly have been without some such system of computation. The Quetzalcoatl faith, however, was manifestly of a considerably higher status than that which it encountered, as is obvious not only by the numerous and extraordinary traditions regarding the Toltec civilization, but the actual remains it has left. It is clear that, whether it found a calendar or time-count already existing, it placated aboriginal opinion by the amalgamation of the several festivals of the rain-god with its own. The fact that the day-signs of the Mexican calendar or tonalamatl are almost identical with those of the Maya tonalamatl is good proof that the former was developed from the latter; and if only a small proportion of Toltec deities find a place in its monthly festivals, that would seem to be due to the circumstance that later Nahua demands for the inclusion of their tribal deities were acceded to. We may, perhaps, imagine the early tonalamatl of the Quetzalcoatl missionaries to have been similar in form to that of the Maya—that is, it must have been almost wholly concerned with the festivals of deities of a purely agricultural kind.