^{page 573 note 1} Khiṭaṭ, ii, p. 277.

^{page 573 note 2} The other exception is the mosque of al-Azhar, which appears to have been distinguished by the same anomaly. Maqrīzī (op. cit., ii, p. 273), speaking of the dating inscription, says that it was written round the rim of the dome, which was in the first aisle (riwāq) of the sanctuary, and to the right of the miḥrāb and the minbar. It was therefore probably in the back corner, as the first aisle must, of course, be taken to mean the one next the qibla wall. If it was balanced by a second in the left-hand corner, the arrangement would be the same as in the mosque of al-Ḥākim. Rivoira (Moslem Architecture, p. 163) is the only author who appears to have realized that the sanctuary of al-Ḥākim's mosque must have had three domes. As for al-Azhar, the significance of the above-quoted passage in Maqrîzî has escaped notice entirely.

^{page 573 note 3} In Cairo the direction of Mekka is almost exactly south-east.

^{page 574 note 1} “ Notes d'archéologie arabe,” in the Journal Asiatique, viii^e série, tome xvii, pp. 439–441Google Scholar.

^{page 574 note 2} He expressly says that the ramp cut diagonally across the windows and that a great part of the Kufic inscriptions were hidden by it.

^{page 575 note 1} See the Comptes rendus du Comité de Conservation, 1913, p. 106.

^{page 575 note 2} van Berchem, C.I.A., i, pp. 103–4.

^{page 575 note 3} See my “Brief Chronology of the Muhammadan Monuments of Egypt” : Bull, de l'Inst. français d'Archéo. orientale, t. xvi, pp. 77–78Google Scholar.

^{page 575 note 4} van Berchem, C.I.A., i, pp. 733–6.

^{page 575 note 5} These finials, on account of their resemblance to censers, are known locally by the name of mab āra.

^{page 576 note 1} The earliest example of this motive that I have seen occurs at Tabga (Capernaum) on the fallen lintel of a second century synagogue, a fact which tends to confirm its traditional Jewish origin. The Tabga example, according to the Jewish Encyclopædia, is the earliest example known. The earliest example of its use as a decorative motive in Islam is that referred to above.

^{page 576 note 2} The ornament of these minarets, and of the rest of the mosque also, has been admirably treated by Flury, in his Ornamente der Hakim- und Ashar-Moschee, Heidelburg, 1912Google Scholar.

^{page 576 note 3} Khiṭaṭ, ii, p. 278, 1. 9 f. (translated by van Berchem, “ Notes d'archéologie arabe,” loc. cit., pp. 435–6).

^{page 577 note 1} Ibid., pp. 444–5.

^{page 578 note 1} Prisse d'Avennea, L'Art aràbe, texte, p. 98.

^{page 578 note 2} I say “ military architecture ” as this type of masonry is only found in the three gates and part of the wall of Cairo, but it is not found in the mosque of al-Juyūshy.

^{page 581 note 1} He had already seen it and noted it; op. cit., p. 19, note 43, and his Islamische Schriftbänder, p. 10, note 6.

^{page 582 note 1} In the course of his description of the main artery of Cairo from the Bāb Zuweyla to the Bāb al-Futūḥ (Khiṭaṭ, i, p. 376, and Casanova's transl., MIFAO, tome iv, pp. 78–9).

^{page 582 note 2} For this inscription see van Berchem, Notes d'archéologie arabe, in the Journal Asiatique, 8^me série, t. xvii, pp. 455–60, and his C.I.A., i, pp. 61–2.

^{page 583 note 1} Op. cit., ii, p. 277 (translated by van Berchem, “ Notes,” loc. cit., p. 434). The works of al-Masīḥi are lost.

^{page 584 note 1} Mahmoud Bey, “ Le Systéme métrique actuel d'Égypte,” in the Journ. Asiat., 7° série, tome i, pp. 100–1.

^{page 584 note 2} This measurement is not direct, owing to the buildings clustering round the base of this salient, and therefore it cannot be quite accurate.