[315] Description de l'Égypte, vol. iii. 55.
[316] According to Gau, there was, in 1817, a well preserved tabernacle in the sanctuary of the temple at Debout, in Nubia. (Antiquités de la Nubie, 1821, pl. v. Figs. A and B.)
[317] De Rougé, Notice des Monuments, etc. (Upon the ground floor and the staircase.) Monuments Divers, No. 29. The term naos has generally been applied to these monuments, but it seems to us to lack precision. The Greeks used the word ναός or νεώς to signify the temple as a whole. Abd-el-Latif describes with great admiration a monolithic tabernacle which existed in his time among the ruins of Memphis, and was called by the Egyptians the Green Chamber. Makrizi tells us that it was broken up in 1349. (Description de l'Égypte, Ant., vol. v. pp. 572, 573.)
[318] Herodotus, ii. 175.
[319] Translated by Maspero, Histoire Ancienne, p. 385. The whole inscription has been translated into English by the Rev. T. C. Cook, and published in vol. ii. of Records of the Past.—Ed.
[320] As M. Maspero has remarked (Annuaire de l'Association des Études Grecques, 1877, p. 135), these secret passages remind us of the movable stone which, according to Herodotus (ii. 121), the architect of Rhampsinit contrived in the wall of the royal treasure-house which he was commissioned to build. Herodotus's story was at least founded upon fact, as the arrangement in question was a favourite one with Egyptian constructors.
[321] Ἀναγλυφὰς δ' ἔχουσιν οἱ τοῖχοι οὗτοι μεγάλων εἰδώλων (Strabo, xvii, 1, 28).
[322] Description de l'Égypte, Antiquités, vol. i. p. 219. The authors of the Description générale de Thèbes noticed recesses sunk in the external face of one of the pylons at Karnak, which they believed to be intended to receive the leaves of the great door when it was open (p. 234); they also noticed traces of bronze pivots upon which the doors swung (p. 248), and they actually found a pivot of sycamore wood.
[323] These measurements are taken from Mariette, Voyage dans la Haute-Égypte, vol. ii. p. 7.
[324] We have not given a general map. In order to do so we should either have had to overpass the limits of our page, or we should have had to give it upon too small a scale. Our fourth plate will give a sufficiently accurate idea of its arrangement. The plan in Lepsius's Denkmæler (part i. pl. 74-76) occupies three entire pages.