Page:History of Art in Sardinia, Judæa, Syria and Asia Minor Vol 2.djvu/130

 114 A History of Art in Sardinia and Jud^a. door at the side (2 in plan). These apartments are all of different sizes ; but to the west a large chamber interposes between a set of small cells, probably occupied by the household. This part of the building was doubtless the male quarter, or selamlik, whilst the harem must be sought in the after-part of the edifice, beyond the door at the end of the court. The narrow opening in the east wall (3 in plan) may have belonged to it, as well as a chamber close by, with no outlet save towards the court, which we must consider the room reserved to eunuchs. However this may be, there is no doubt as to five or six apart- ments, three of which are rather large, forming a block by itself at the end and side of the court. Two of these chambers (5 and 7 in plan) have rock excavations of unequal size, probably intended as bath-rooms (6 and 8 in plan) ; a depression or tank in the main court (9 in plan) should also be noticed. It may have served to collect rain water, brought about by a slight incline in the pave- ment. Diggings would probably reveal the channel excavated through the rock for draining purposes. The decoration about the throne, and a cavetto moulding on either side of the foundation wall, show that the chisel of the sculptor had been freely employed to beautify a building whose proportions, with its thirty apartments, great and small, make it clear that it was the most important in the place. The general disposition of the edifice is akin to that of the palaces in Chaldaea, Assyria, and Jerusalem. On the supposition of its having been a temple, it is not easy to account for, nor what use could be assigned to the series of small apartments ranged around the central court, whilst there is no chamber sufficiently distinct from the others to have been the habitation of the deity. A similar temple would have had none of the characteristics of those we have described, or shall describe in a future work. On the other hand, all its features — the terraces, for instance — seem to have been calculated for a domestic and not a reliofious dwelling. Fronting the palace, at a height of five metres from each other, were two esplanades, carefully laid out, each measuring 140 m. in length by no m. in width. ^ Leading to them, on either side of the hill, were two magnificent flights of steps, indicated on the ^ The measurement is taken from Texier's plan, and within a few yards coincides with reality.