Page:History of Art in Phœnicia and Its Dependencies Vol 1.djvu/384

362 HISTORY OF ART IN PHOENICIA AND ITS DEPENDENCIES. were store-rooms ; elephants and horses would not, he says, be lodged on the abrupt slopes of the acropolis. Hut perhaps the most probable explanation of chambers like these, lying upon the rock and all communicating with each other, is to suppose that they were cisterns or reservoirs. 1 It would be easy to keep them always full, for the catching surface at command was great, and nothing but a good system of pipes and channels was required for its proper utilisation. Such a precaution seems to have been universal in Punic fortifications ; this same arrangement has been found at Adrumetum, at Utica, at Thapsus, and at Thysdrus. In .this respect foresight was carried so far that even the second line of defence, standing some forty to fifty yards in front of the great rampart, was supplied with similar chambers (Fig. 249). The mercenaries who formed the garrison thus had their own supply of water beneath their feet and did not need to encroach upon the resources of the townsfolk. From all these facts and considerations we may gather the following general idea as to the constitution of the great rampart of Carthage. Above the cisterns hidden in its foundations the wall must have been practically solid for a considerable height, that is to say, up to above the highest point to which a battering-ram could reach. There was nothing, however, to forbid the erection of stables for horses and elephants immediately behind the rampart. Above the solid part of the wall there were chambers, either vaulted or ceiled with timber, in which soldiers could be lodged and war material stored. There may have been one or two rows, or one or two stories of these chambers, as Appian tells us, and their arrangement may have varied in order to fit the trace of the wall. Their front walls must have been very thick, and pierced with loopholes. Above them ran the barbette. At regular distances of two plethra, or 206^ feet, rose the square towers with which the wall was flanked. 2 Being higher than the curtain by two stories they enabled the defenders to pour missiles on the flank of an assailant even after he had reached the summit of the wall, while they afforded a post of vantage for artillery. 8 1 DAUX, Recherches, pp. 190-192. On this point GRAUX is of the same opinion as Daux. Note, p. 196. 2 It is from APPIAN (viii. 95) that we get this distance of two plethra for the intervals between the towers ; he also tells us that the towers were four stories high. 3 DAUX, Recforches, pp. 193, 194.