of the corn are not infrequently represented in the form of horses; and, second, that the animal which in later legends is said to have injured the god was sometimes originally the god himself, we may conjecture that the horses by which Virbius or Hippolytus was said to have been slain were really embodiments of him as a deity of vegetation. The myth that he had been killed by horses was probably invented to explain certain features in his worship, amongst others the custom of excluding horses from his sacred grove. For myth changes while custom remains constant; men continue to do what their fathers did before them, though the reasons on which their fathers acted have been long forgotten. The history of religion is a long attempt to reconcile old custom with new reason, to find a sound theory for an absurd practice. In the case before us we may be sure that the myth is more modern than the custom and by no means represents the original reason for excluding horses from the grove. From their exclusion it might be inferred that horses could not be the sacred animals or embodiments of the god of the grove. But the inference would be rash. The goat was at one time a sacred animal or embodiment of Athena, as may be inferred from the practice of representing the goddess clad in a goat-skin (_aegis_). Yet the goat was neither sacrificed to her as a rule, nor allowed to enter her great sanctuary, the Acropolis at Athens. The reason alleged for this was that the goat injured the olive, the sacred tree of Athena. So far, therefore, the relation of the goat to Athena is parallel to the relation of the horse to Virbius, both animals being excluded from the sanctuary on the ground of injury done by them to the god. But from Varro we learn that there was an exception to the rule which excluded the goat from the Acropolis. Once a year, he says, the goat was driven on to the Acropolis for a necessary sacrifice. Now, as has been remarked before, when an animal is sacrificed once and once only in the year, it is probably slain, not as a victim offered to the god, but as a representative of the god himself. Therefore we may infer that if a goat was sacrificed on the Acropolis once a year, it was sacrificed in the character of Athena herself; and it may be conjectured that the skin of the sacrificed animal was placed on the statue of the goddess and formed the _aegis,_ which would thus be renewed annually. Similarly at Thebes in Egypt rams were sacred and were not sacrificed. But on one day in the year a ram was killed, and its skin was placed on the statue of the god Ammon. Now, if we knew the ritual of the Arician grove better, we might find that the rule of excluding horses from it, like the rule of excluding goats from the Acropolis at Athens, was subject to an annual exception, a horse being once a year taken into the grove and sacrificed as an embodiment of the god Virbius. By the usual misunderstanding the horse thus killed would come in time to be regarded as an enemy offered up in sacrifice to the god whom he had injured, like the pig which was sacrificed to Demeter and Osiris or the goat which was sacrificed to Dionysus, and possibly to Athena. It is so easy for a writer to record a rule without noticing
