Page:Archaeological Journal, Volume 8.djvu/316

 230 SILCH ESTER. For though the one from north to south runs directly from one gate to the other, as drawn in the plan, the other does not run directly from east to west, as is stated ; and if it did, the streets could none of them be at right angles to each other, which in fact they are ; the eastern street being a continuation of the Roman way from the eastward, through the eastern gate to the forum, or centre building ; and the western street running in the direction of the south-east angle of the work, and passing on continuously by the north end of the centre building. It would be necessary to examine these streets year after year, as the crops come on successively, in the way Mr. Stair did, to be able to make out the whole of them ; but they have been sufficiently examined to show that the principal streets were towards the true cardinal points, and conse- quently at right angles to each other. ^ Such as have been observed on the ground are introduced on the map in dotted hues. A minute account of the wall will be found in the " Philo- sophical Transactions," No. 490, a.d. 1748. It appears to have been about 13 feet high, and about 8 feet thick at the bottom, composed of layers of flat stones about 30 inches apart, with flints between them, set in very strong mortar.^ The circuit of the wall is about a mile and a half, and the area inclosed is about 102 acres. The exterior line of defence, which is at an irregular distance, averaging 170 yards from the wall, consists of a rampart and outside ditch, which, when complete, may have been continuous all round, but at present there is no reason to suppose it to have been carried round on the south-east side.® The height of the rampart of this exterior line seems ^ " The interior has long been subject every direction, that at the earnest re- to the action of the plough ; but to the quest of the tenant, the proprietor desired eyes of an antiquary the directions of the that the foundations should be covered in. ancient streets, at right angles to each See the position of the bath, as laid dow-n other, are yet perceptible, by a difference in Mr. A. J. Kempe's Map. (Archseologia, in the height of the corn growing on them vol. xxvii., p. 419. Plate xxxii. Ap- when compared with its general surface." pendix.) (United Service Journal, January, 1836, ^ We could not see the part of the wall p. 38.) whei-e, it is stated, the flat courses were The position of the bath has been fixed six in number ; Jive seems the most com- on the map by the concurrent testimony nion number of horizontal courses, making of three persons residing at Silchester, about 1 3 feet. who saw the excavation open. Though the " The value of the ground, and its rector had carefully fenced in a way to southern aspect, may perliaps account for the remains, so as to protect the farmers the more complete destruction of the crops, such was the destmction coin- outer entrenchment on the southern side, mitted by persons crossing the fields in