consider Justin’s statement hereafter. For the moment we pass on to Heliopolis (Baalbec), where, the ecclesiastical historian Socrates affirms, virgins were offered in prostitution to strangers.[267.2] He does not, any more than Justin, connect this with a temple or a divinity; but from Sozomen we gather that it was a religious observance, inasmuch as the prostitution of virgins prior to their marriage is stated to have been abolished by Constantine when he destroyed the temple of Aphrodite.[267.3] A similar custom, according to Ælian, was followed by the Lydians. And he expressly says that when once the rite had been performed the woman remained ever afterwards chaste, nor would a repetition be forgiven her on any plea.[267.4] Herodotus, however, states that the daughters of the common people in Lydia earned their dowries by a life of prostitution.[267.5] The two writers are obviously referring to two different customs. A third custom distinct from either is mentioned by Strabo as practised by the Armenians, among whom even the highest families of the nation consecrated their virgin daughters to the service of the goddess Anaitis, to remain as prostitutes at her temple before their marriage.[267.6] {268} What is the relation of these three customs? They have usually been considered as closely connected. It may be, as Dr Frazer suspects, that the real motive for the custom described as that of earning dowries by prostitution was religious, rather than economical, although my own suspicions point in another direction. But putting that custom aside for further examination, both the others are certainly portrayed as religious. As practised by the nations of Western Asia for a thousand years prior to the fall of paganism they were annexed to the cult of certain divinities. There is, however, a broad distinction to be drawn between a custom requiring every woman once in her lifetime to submit to the embraces of a stranger, and one which consecrated a life of prostitution. Such a life was one of devotion to the goddess as a more or less permanent servant. The other custom demanded a single act which freed the worshipper for the rest of her days. It may be freely conceded that the goddess at whose temple, or on whose feast-day, the act was performed was endowed with similar characteristics to those of the goddess in whose service the life of prostitution was lived. It may even have been that sometimes the same goddess had bands of harlots attached to her shrine, and also required the sacrifice of the virginity of all other women in the manner described. This perhaps, as we shall see, was the case in Lydia. We should still need to investigate separately the two customs. One of the most fertile sources of error in the interpretation of custom is the fatal tendency of rites distinct, or even altogether different in origin and intention, but similar in expression, to converge. This convergence is accelerated by a variety of causes. The