by a knotted pinnacle and crescent, with several wooden staffs fixed at angles to the round of the cupola, from which lamps are suspended on the great festivals. Two or three galleries project at various heights, supported by stalactite corbels and cornices, and from these the muezzin proclaims the call to prayer five times a day. It is recorded by El-Makrizy that the first stone minaret in Cairo was that of the mosque of El-Māridāny, built by the Master Suyufy-all the earlier ones being of brick.* A very beautiful example of a minaret is seen in the engraving of the mosque of Kaït Bey (frontispiece). Sometimes the cupola at the top is fluted, as in a very pretty little minaret in the southern burial-ground of Cairo, which tapers upwards from the square by a series of diminishing octagons till the transition to the round can be gently effected. The transitions are ingeniously managed by those stalactite or pendentive ornaments, which are the peculiar property of the Saracenic architect, and are freely used to mask angles and to modulate such transitions as those in the dome and minaret. In describing the minaret we are, however, anticipating the true chronological order, for the earlier mosques do not present many of the graceful details which we see in that of Kaït Bey. The great minaret of Ibn-Tūlūn indeed diminishes by stages, but there are no stalactites in any part of this mosque, except over the mihrab, or niche, and these are probably a later addition.

Nothing has been said so far about the dome, and for this reason, that the mosque of Ibn-Tūlūn has none. It is a mistake to suppose that the dome is an essential feature of a mosque. The minaret is essential, because there must be a raised tower from which the Adan or Call to Prayer may resound over the city, though even this was dispensed with in the Prophet's own mosque at Medina, where the Muezzin Bilal of the stentorian

* El-Māri āny's mosque is well illustra'ed in Ebers' Egypt, ii. 70; and the minaret is separately engraved in i. 61. It is converted from the square into an octagon very near the base, and thence at the first stalactite gllery into the round; above the second gallery (there are but two) is a stone neck or p nnacle, twelve courses high supporting a conical bulb-like crown.

voice shouted the call from the gate. A dome, however, has nothing whatever to do with prayer, and therefore nothing with a mosque. It is simply the roof of a tomb, and only exists where there is a tomb to be covered, or at least where it was intended that a tomb should be. Only when there is a chapel attached to a mosque, containing the tomb of the founder or his family, is there a dome, and it is no more closely connected with the mosque itself than is the grave it covers: neither is necessary to the place of prayer. It happens, however, that a large number of the mosques of Cairo are mausoleums, containing chambers with the tomb of the founder, and the profusion of domes to be seen, when one looks down upon the city from the battlements of the Citadel, has brought about the not unnatural mistake of thinking that every mosque must have a dome. Most mosques with tombs have domes, but no mosque that was not intended to contain a tomb ever had one in the true sense. The origin of the dome may be traced to the cupolas which surmount the graves of Babylonia, many of which must have been familiar to the Arabs, who preserved the essentially sepulchral character of the form, and never used it, as did the Copts and Byzantines, to say nothing of European architects, to roof a church or its apse. The form of the true Cairo dome is not quite the same as that of Italy and St. Paul's; like most Saracenic designs it is based upon simple geometrical proportions. To draw the outline of the ordinary type (fig. 6), to which, however, there are exceptions, describe a circle A, draw tangents B B, to the length of three-fourths of the radius, join the extremities, and from each of the extremities draw a circle c, the radius of which shall equal the whole diameter of the first circle plus an eighth; and where these circles intersect erect the pinnacle. The whole can be done with compasses and rule.

Domes are generally built of brick, not moulded to fit the curve, but simply laid each tier a little within the lower tier so as to form the proper curve; the plaster which coats most don es

inside and out conceals the slight irregularity of the brickwork. Wooden frames are also sometimes used to support the lighter plaster domes, as is shown in the foreground of fig. 4. Some domes, however, are of stone, which is cut to the shape of the curve, and carved with the desired pattern. As a rule I have observed that plain and fluted domes are of plastered brick, whilst those ornamented with zigzag, geometrical, and arabesque devices are more commonly of carved stone. The surfaces of the domes

are ornamented in various ways. Sometimes they are covered with an intricate geometrical design, with star centres, as the domes of Kaït Bey and Al-Ashraf Bars-Bey in the eastern cemetery. A common decoration consists in bands of zigzags, or chevrons close together, running horizontally round the dome from base to apex, such as we see in the tomb-mosque of Barkūk (1407). Many domes are fluted, and these would seem to belong to all periods of Cairo architecture, for we find the fluted cupola surmounting the mibkharas or quasi-minarets of the mosque of El-Hakim (1012; but these may belong to the

restoration, in 1303, when it is known that the mibkharas were shored up with massive bases), and also in domes in the southern burial ground, which apparently belong to the end of the 15th century. A rarer and late form of dome ornament consists in covering the whole surface with arabesques arranged in large outlines, which form a sort of diaper, with a much richer effect than mere geometrical ornament. There are a few

examples, which are probably of very early date, with a lantern pierced with small windows, and roofed with a little fluted cupola on the top of the larger dome. These are in the southern burial-ground, but are in so ruined a condition that there remains no evidence as to their date that can be regarded as positive. Certain characteristics of the stalactites, however, lead to the belief that they may belong to the Ayyuby period (1170-1250). Some of the more elongated domes have a second and lower dome structure inside them, from which spring