i8o A History of Art in Sardinia and Jud^a. simple aperture. Beneath the floor, a double chamber, now entered by a side hole, a late addition that did not belong to the original scheme, had also been scooped out. The section (Fig. 347) shows the commencement of another sepulchral memorial, which was left unfinished, as well as the talus of the hill, on the face of which the column was cut. It will be seen that the rule laid down by Fig. 345.— Plan of Gherdek-Kaiasi. Plate XXXII. •rrvtiri ^u u^ Fig. 346. — Gherdek-Kaiasi. Longitudinal Sec- tion under the Porch. Plate XXXIII. Vitruvius ^ was strictly adhered to, the weight being thrown on the exterior line. It is not easy to date this monument, for it may with equal propriety have been due to Hellenic influence, as it may be a primitive type borrowed by Asiatic Greeks. Pillars belonging to early Grecian art, and closely resembling these, adorn the rock-cut tombs at Cyrenae ; ^ in them, however, nor in the monuments of Assyria, Egypt, and Phoenicia, so far as we know, were antae introduced — at least, not in the way they are in this instance. On the other hand, the absence of frontel, frieze, or cornice about this portico, recalls Oriental arrangement and methods. Thus an insig- nificant capital alone interposes between the pillar and the entablature, which is formed of a double fascia extending along the whole porch. The lower band is a mere outline ; whether this was in the original plan, as in the Persepolltan tombs, or brought about by exfoliation of the rock cannot be determined. The capital has none of the features proper ^ Tom. iii. chap. v. Doric columns are not bounded by "vertical" lines, as was at first assumed, and as stated in the text. It is now found that they have a convex profile.— Editor. ^ Consult Pacho, Voyage dans la Cyrenaiqjie. Fig. 347. — Transverse Sec- tion, Gherdek - Kaiasi. Plate XXXIII.
