1. The absence of historical foundation for the tradition.
2. The close resemblance between the tradition and other stories and superstitions which unquestionably deal with heathen goddesses, such as Berchta and Hertha.
3. The equally close analogy between the procession and that described in Eastern stories, which, so far as we know, could not have reached England at the latest period when the procession could possibly have been instituted; and between the procession and certain heathen rites practised not only in the East, but as near home as Rome and Germany,—nay, in Britain itself.
4. The occurrence of a similar procession at Southam, in the same county, having the special feature of a black lady, best explained as a survival of certain rites practised by the ancient Britons.
5. The connection between the analogous legend at St. Briavel's and the remains of a sacred communal feast that can hardly be anything else than the degraded remnant of a pagan observance.
The want of historical evidence cannot, of course, be overlooked; but we must remember that in investigating traditions and traditional observances we are dealing with a phase of civilization of which history only yields rare and indirect glimpses. It is the absence of direct evidence that, not only in the science of Folklore, but also in the physical sciences, causes resort to the evidence afforded by comparison of other structures and processes. On the validity of this evidence, and -the reasoning based upon it, nearly all our scientific learning depends. In spite, therefore, of the defects in the historical evidence, and in the absence of evidence to the contrary, it can scarcely be denied that the analogies in both custom and legend here brought together amount to a fairly strong presumption in favour of the conclusions I have ventured to draw from them.
If I may formulate my conjecture as to the course of development actually pursued, it would be something like
