Mosques are even more notable than private houses for the inscriptions on their walls. Every visitor to St. Sophia remembers the great green medallions bearing the names of the chief personages of Islam in letters of gold. In purely Turkish mosques similar medallions may be seen, or large inscriptions stencilled like panels on the white walls, or small texts hanging near the floor. But there is a more architectural use of writing, above doors and windows or in the form of a frieze. When designed by a master like Hassan Chelibi of Kara Hissar, the great calligrapher of Süleïman’s time, and executed in simple dark blue and white in one of the imperial tile factories, this art became a means of decoration which we can only envy the Turks. Such inscriptions are always from the Koran, of course, and they are often happily chosen for the place they occupy. Around the great dome of the Süleïmanieh, and lighted by its circle of windows, runs this verse: “God is the light of the heavens and of the earth. His light is like a window in the wall, wherein a lamp burns, covered with glass. The glass shines like a star. The lamp is kindled from the oil of a blessed tree: not of the east, not of the west, it lights whom he wills.” It is not only for inscriptions, however, that tiles are used in mosques. Stamboul, indeed, is a museum of tiles that has never been adequately explored. Nor, in general, is very much known about Turkish ceramics. I suppose nothing definite will be known till the Turks themselves, or some one who can read their language, takes the trouble to look up the records of mosques and other public buildings. The splendid tiles of Süleïman’s period have sometimes been attributed a Persian and sometimes a Rhodian origin—for they have many similarities with the famous Rhodian plates. The Turks themselves generally suppose that their tiles came from Kütahya, where a factory still produces work of an inferior kind. The truth lies between these various theories. That any number of the tiles of Constantinople came from Persia is impossible. So many of them could not have been safely brought so far overland, and it is inconceivable that they would have fitted into their places as they do, or that any number of buildings would have been erected to fit their tiles. The Rhodian theory is equally improbable, partly for similar reasons though chiefly because the legend of Rhodes is all but exploded. The Musée de Cluny is almost the last believer in the idea that its unrivalled collection of Rhodian plates ever came from Rhodes. Many of them probably came from different parts of Asia Minor. That tiles were produced in Asia Minor long before the capture of Constantinople we know from the monuments of Broussa, Konia, and other places. They were quite a different kind of tile, to be sure, of only one colour or containing a simple arabesque design, which was varied by a sort of tile mosaic. Many of them, too, were six-sided. The only examples of these older