[216] Keramoi d' ou chrontai, says Strabo. These words, as Letronne remarked _a propos_ of this passage, combine the ideas of a tiled roof and of one with a ridge. The one notion must be taken with the other; hence we may infer that the Babylonian houses were flat-roofed. [217] STRABO, ii. 5, 11. [218] See M. AMEDEE TARDIEU'S reflections upon Strabo's method of work, in his _Geographie de Strabon_ (Hachette, 3 vols, 12mo.), vol. iii. p. 286, note 2. [219] As to this singular people and their religious beliefs, the information contained in the two works of Sir H. LAYARD (_Nineveh_, vol. 1. pp. 270-305, and _Discoveries_, pp. 40-92) will be read with interest. Thanks to special circumstances Sir H. Layard was able to become more intimately acquainted than any other traveller with this much-abused and cruelly persecuted sect. He collected much valuable information upon doctrines which, even after his relation, are not a little obscure and confused. The Yezidis have a peculiar veneration for the evil principle, or Satan; they also seem to worship the sun. Their religion is in fact a conglomeration of various survivals from the different systems that have successively obtained in that part of Asia. They themselves have no clear idea of it as a whole. It would repay study by an archaeologist of religions. [220] BOTTA, _Monument de Ninive_, vol. v. p. 70. [221] See above, page 118, note 1. [222] Some rooms are as much as thirty feet wide. They would require joists at least thirty-three feet long, a length that can hardly be admitted in view of the very mediocre quality of the wood in common use. [223] _Gailhabaud, Monuments anciens et modernes_, vol. i.; plate entitled _Tombeaux superposes a Corneto_. [224] PLACE, _Ninive_, vol. i. p. 309. In this passage M. Place affirms that Mr. Layard discovered in a room of one of the Ninevite palaces, several openings cut at less than four feet above the floor level. It is, moreover, certain that these openings were included in the original plan of the building, because the reliefs are interrupted so as to leave room for the window without injury to the scenes sculptured upon them; but, adds M. Place, this example is unique, one of those exceptions that help to confirm a rule. We have in vain searched through the two works of Sir Henry Layard for the statement alluded to by M. Place. The English explorer only once mentions windows, and then he says: "Even in the rooms bounded by the outer walls there is not the slightest trace of windows" (_Nineveh_, vol. ii. p. 260). [225] BOTTA, _Monument de Ninive_, vol. v. p. 73. [226] FLANDIN et COSTE, _Voyage en Perse; Perse ancienne_, plates 28 and 29; and, in the text, page 25. These openings occur in the great Sassanide palace at Ctesiphon, the _Takht-i-Khosrou_ (_ibid._ pl. 216, and text, p. 175). Here the terra-cotta pipes are about eight inches in diameter. According to these writers similar contrivances are still in use in Persia.