that theory, particularly as it failed to account for the openings in the circles, the absence of figures indicating the sun and moon, and not even the rudest attempt at the phases of the lunar body. It was also suggested that the circles were intended for moulds to cast metal rings. The fact of the circles so often occurring on a sloping face of a rock renders this untenable. That the circles were used for some game was rejected from the varying sizes of the circles on different stones. ‘The idea which occurred to my own mind,’ he continues, ‘was, that the incised circles were intended to represent the circular buildings of earth or stone, of which the traces still exist in every part of Ireland. This conjecture is supported by the following considerations:--1. The circles are of different sizes, and some are disposed in concentric groups. The ancient dwellings and fortified seats of the ancient Irish were circular; they were of various sizes, from the small cloghan, or stone house, of ten feet in diameter, to the great camp including an area of some acres; and the principal forts had several concentric _valla_. 2. The openings in the inscribed circles may have been intended to denote the entrances. 3. The other inscribed lines may have represented roads passing by or leading up to the forts.’ Discouraged by the reception with which his theory was received, he laid his drawings aside for many years. Returning again to the subject, he gave it further examination, which confirmed his original opinion. He says: ‘The centres of the circles and the neighbouring cups and dots arrange themselves generally three by three in straight lines. This disposition of the symbols could not be said to be perfectly accurate; but I thought I could observe close and designed approximation to it. If, then, the circles represent forts, and are disposed three by three in straight lines on the inscribed stones, I saw that we might expect to find the forts disposed in like manner over the surface of the country; and I think that I have succeeded in verifying this inference. The ancient raths have fortunately been laid down on the six-inch Ordnance Survey maps of Ireland; and unless I am deceived by fortuitous collineations, I find that the forts are actually arranged three by three in straight lines. The discovery of this fact, if it be a fact would be of much more consequence than the explanation of the meaning of the inscriptions of which I have just given an account. But this further inquiry must be conducted with care. Large portions of the country must be examined, and those difficulties must be confronted which the disappearance of ancient remains must inevitably give rise to.’[11]