quently to Roman times, and generally as belonging to that transitional age when Christianity was struggling with Paganism for the mastery. The common people are generally willing enough to amuse themselves with fables about giants and demigods, and to wander back into prehistoric times; but with regard to these monuments they do not seem to have done so. I do not recollect a single tradition that ascribes any stone circle to the pre-Roman period.
If, however, I am correct in assuming that these great groups of circles belong to the Arthurian age, we have no difficulty in assigning to this one its proper place in the series of his battles. The ninth, as we have seen above, was probably fought at Caerleon on the Usk; which would seem to indicate that, at a certain point in his career, Arthur was forced back quite out of England into South Wales; but his return on that hypothesis is easily traced. The tenth battle was on the shore of some large river, which ought in consequence to be the Severn, though the name given in the text lends no countenance to this supposition; the eleventh was "In monte quod dicitur Agned in Somersetshire," which would answer perfectly, except in name; for Stanton Drew, in that case, would be in the direct line of advance to Badon Hill, where the twelfth and crowning victory was fought.
The name here, as throughout, creates the difficulty, but Stanton on the Stones, or Stone Town, is simply an epithet applied to all these groups by the Saxons at some period subsequent to that of which we are speaking, when the memory of their purpose was lost, or little cared for by those of a different race, and speaking a different language, who had succeeded to the Bryts, who had erected them. Unless we assume that Stonehenge, Stanton Drew, the circles on Stanton Moor, and the stones at Stennis, and others, were erected by the Saxons themselves, they must originally have borne Celtic names, and it would be these names that Nennius would quote, and which consequently could not be those by which they are now known.
The expression "in monte" is singularly confirmatory of this determination, inasmuch as one of the remarkable features of the
