^{page 759 note 1} To avoid confusion, I have adopted throughout the terms of orientation as given byAl-Makrīzy. He treats the Kibla side of the Mosque (i.e. that supposed to be directed towards Mekka) as south; and names the other sides accordingly. As a matter of fact, the axis of the mihrāb points as near as possible S E by E.: and therefore it would of course be nearer to the facts to call it E.; but the confusion it would introduce in following our authority through the history would be considerable, while real accuracy would not be obtained. The plans are therefore all marked according to Al-Makrīzy's system, and all references to the points of the compass are to be understood in this sense. The correct orientation is also indicated outside Plan I.

^{page 761 note 1} This plan is reduced from one drawn by an official of the Wakfs in 1873, and liberally placed at my disposal by the said department. It may not be absolutely correct; but it is evident that considerable care has been spent upon it; and it is, in my own belief, the first and only plan of the Mosque that has even approached to correctness. Those of Coste, and of the various guide-books, represent the Mosque as really a rectangular figure; and though sufficing to show the general scheme, are worse than worthless as a basis of investigation.

^{page 762 note 1} Of all the columns in the E and W. colonnades the bases only remain; but that is sufficient for our purpose.

^{page 764 note 1} The date of the flight of the Prophet to Al-Medīna, from which the Muslims date their era, corresponds to July 16, A.D. 622. Our authorities of course give their dates according to this era, which will he found much the most convenient for us to use in tracing the history, as each date will at once suggest the age of the Mosque, which is only twenty-one years younger than the era. I have therefore given the dates in years of the Flight (4.H.); but I have also in each case given the corresponding date A.D.

^{page 766 note 1} The name of the cubit is in the Arabic dhirā-al-amal. Now there are and have been many different cubits in use in Egypt; but no cubit of this name is now known, nor have I been able to discover any account of it. If, however, we look through the list of cubits now known, that called the balady or “native” cubit at once suggests itself as the most likely to be meant by dhirā-al-amal, which expression would seem to mean “the cubit of ordinary use.” Further, we shall see later, that if we accept this identification, the result is an almost absolute coincidence in the length of the kibla side of the Mosque of the present day, with the recorded length of that side after the last extension in that direction of which any account has come down to us. I therefore take the dhtrā-al-umal to be the same as the dhu ā balady, or 578 metres.

^{page 766 note 2} Here ends the testimony of the eye-witness.

^{page 766 note 3} For we hear afterwards that Maslama first spread mats instead of the former pebbles, A.H. 53 (A.D. 673).

^{page 766 note 4} The Mihrāb itself is often vulgarly called Kibla, but not with strict correctness. Kibla simply means “direction towards,” and technically the direction of Mekka taken by the Muslim in prayer.

^{page 767 note 1} For we learn further on that the walls were first plastered by Maslama, 32 years later.

^{page 768 note 1} There is no reason, so far as I can see, to doubt the essential authenticity of this discourse, which I translate from Al-Makrīzy, ii. 260. It is related by Ibn-Lahaia, who was born A.H. 89; and the tradition is traced through two other persons, to the earlier of whom the first-hand witness, Buhair-ibn-Dhākir, related as in the text. For the better understanding of the sermon, it is necessary to remark that in the early days of Islām in Egypt no Muslims were allowed to settle in the country districts; they were confined by the Caliph's special order to Fustāt and Alexandria. In the spring however, the great body of the armbearing population went to camp for three or four months in different parts of the country, for the purpose of putting out their horees to grass. The custom exists, in a necessarily modified form, to this day; and the technical term for it is the same that was used in Amr's time. Any native gentleman considers it necessary that his saddle-horses should spend at least three months in the cloverfields, say from the middle of January to May, and in some districts one may see at this season a large plain of clover dotted over with little huts made of maizestalks, where the attendants of the horses sleep.

^{page 768 note 2} I take the Arabic here to mean the Festival of the Baptism of Christ, known among the Copts as Id al Ghitās, the “Festival of the Plunging.” The expression actually used in this passage, and signifying “The Bathing of the Christians,” seems to be unknown to anybody at the present day, but can hardly, I think, refer to anything else. As this festival falls in the middle of January, it just coincides with the period when the horses are put out to grass.

^{page 772 note 1} Our authority says definitely that Abd al Azīz knocked down the existing mosque. This is sufficient to dispose once for all of any attempt to discover even a fragment of the original building in the present Mosque. We shall see later that there can really be but little of the Mosque which has not been lebuilt many times.

^{page 773 note 1} It is doubtful whether Amr had set up another mimbar after the death of the Caliph Omar: it is certain, however, that one already existed in the Mosque at the time of Kurra's additions. According to one account it dated from the time of Abd al Azīz ibn Marwān (A.H. 65–86; A.D. 685–705), and came from a Christian church: according to another, it was sent to the governor Abdallah ibn Sad, the immediate successor of Amr in the governorship of Egypt, A.H. 24–35 (A.D. 645–656), by the King of Nubia, who sent his carpenter, by name Buktur, a native of Dendera, to put it up. Now Buktur is a Coptic name, and the traditions are worth mentioning because they both point to the Coptic origin of the first pulpits in Egypt, and tend to support the theory which refers to Coptic influence that peculiar style of minute and beautiful panelling in wood and ivory, BO characteristic of Muslim pulpits.

^{page 775 note 1} The dhirā al amal (for which see note supra) we will now for convenience call a cubit Simply.

^{page 776 note 1} The calculations are in detail as follows:

E. and W. Colonnades.

The length of the E. and W. sides of the Mosque ia 150 cubits. From this we must subtract cubits, the depth of the S. and N. colonnades, to arrive at the

The Sahn, or court in the middle.

The length and breadth is given by the above:

From N. to S. it is the same as the E. and W. colonnades, viz. cubits.

From E. to W. it ia cubits.

The area of the sahn is therefore , which works out correctly to 5500 aq. cubita.

^{page 777 note 1} That is, always supposing that Ibn al Mutauwag's details are to be relied upon. The fact that the number of columns is really too great to get into the area of his colonnades, together with the coincidence with Pocock's plan just mentioned, inclines one to suspect that the area is too small, and that the Mosque had already, when it contained 378 columns, received the extension N. which we find in Pocock's plan. We shall see later on that there is a further and more important reason for this suspicion.

^{page 779 note 1} The Mosque of Ibn Tūlūn has three ziādas, just as represented in our Plan III.: and a street which runs outside one of them is known to this day as Sikkat az-Ziāda, or Ziāda Street.

^{page 772 note 2} The existence of the eastern Ziāda in later times is proved by the fact that we are told that at some unspecified date after A.H. 702 (A D. 1302), the northern and eastern Ziādas fell down.

^{page 781 note 1} This does not necessarily imply the rebuilding of the S. wall. It would seem that the entire renewal of the S. colonnade is all that is meant: though the wall would of course undergo repairs.

^{page 782 note 1} Ibn Saīd al Maghriby arrived in Alexandria in A.H. 639 (A.D. 1241), and seems to have remained in Egypt about ten years. He visited and examined carefully everything worth seeing in Al-Kāhira and Al-Fustāt, and described them in a book, of which we unluckily possess such fragments only as have been quoted by later authors. The present passage is part ot a longer one quoted by Al-Makrīzy, i. 341 sq.

^{page 783 note 1} In this passage the Sahn is not meant to be included, as we shall see further on. Saladin had pared the S. colonnade, and Aibak would seem to hare repaired this and paved the three others.

^{page 784 note 1} The mihrāb on the outer side, and some remains of blind windows in stucco on the inner side.

^{page 785 note 1} It is to be remarked, that Al-Makrīzy evidently had no notion of the coincidence of Ibn al Mutauwag's square measures with the long measures of A.H. 212.

^{page 786 note 1} The ornaments in the spandrils of the mihrāb are exactly repeated in a frieze running along the S. end of the great Mosque of Baibars (“Fort Schulkowski”) outside Cairo to the North. Altogether, we may take the N. wall as fixed to the time of Baibars.

^{page 786 note 2} Here, again, the actual wall does not seem to be included iu the part destroyed.

^{page 787 note 1} My authority is Aly Pasha Mubārak, now Minister of Public Instruction in Egypt, ia his book Al-Khitat Al-Gadīda, iv. 8, where he quotes from a work entitled Nuzhat an-Nāzirīn, of which I know nothing.

^{page 788 note 1} It will be enough to give the following examples:‘mdash;

“Plan and Upright of a Roman Castrum at Old Cairo” vol. i. pi. ix., which has been characterized by Mr. Butler (Coptic Churches of Egypt, i. 176) as “so very erroneous in places where it can be challenged that it is quite untrustworthy where it cannot.”

“Plan and Upright of the Gate Nasr” (vol. i. pi. xiii.), in which the plan given is really a very incorrect one of the Bāb al Futūh, and not the Bab an Nasr: while the “upright” bears only the very faintest resemblance to either one or the other.

“Map of the Country about Cairo” (vol. i. pi. vi.), where, not to speak of the utter inaccuracy of the whole map, the outline of Cairo itself is quite comically unlike what it has ever been.

“Upright of the Great Pyramid” (vol. i. pi. xvii.), which is at once recognized by the most casual obsei ver to be much too acute an angle.

It is fair to add that in the Preface (p. v) the author himself says: “Nor does he intend to be entirely infallible in his plans: and though he took great pains in measuring all the ancient buildings, yet he is sensible there may be omissions and mistakes, though he endeavoured as much as possible to aviod them.” I need hardly add, that in insisting on Pocock's defect in this respect, I have no thought of detracting from the general merits of that excellent traveller.

^{page 788 note 2} The measurement of A.H. 212, which is only a fraction of a metre more than that of the present day.

^{page 789 note 1} This entertaining and valuable book, the first two volumes of which contain also much information concerning the personages and events of the preceding century, was published at Būlak in A.H. 1297 ( = A.D. 1880). The passage concerning the restoration of the mosque will be found in vol. iii. p. 170.

^{page 790 note 1} The passage in Al-Gabarty was written in 1801, the year of Murad Bey's death.

^{page 790 note 2} The expression is inexact, and therefore we cannot say for certain whether it is meant that the E. wall actually fell. Probably in the first part of the sentence, he is thinking of the S. colonnade, which is always the most important; and in that case the latter part need not apply to the wall, but to the columns only, as the expression would more naturally imply. As a matter of fact, there is only one point in the E. wall (just N. of the northernmost of the two buttresses near the S. corner) where we remark remains of work of the oldest type.

^{page 791 note 1} The new lunar year of the Muslims, 1213, having begun on June 15, 1798.

^{page 792 note 1} The plans in the guides (following Coste) place the Hanafīya in the exact centre, whereas it is a long way nearer to the N. than to the S. side.

^{page 793 note 1} The stucco mihrāb on the outside of the N. wall is in itself sufficient to prove the former existence of a Ziāda here.

^{page 795 note 1} I have since this been informed by the Sheikh Osman Madūkh, a trustworthy authority, that he remembers the Istiska prayer's taking place when he was a small boy, some time ia the fifties,—i.e. between 1250 and 1260 A.H. = A.D. 1834–1844.

^{page 796 note 1} Aly Pasha Mubārak, Al Khitat al Gadīda, vol. iv. p. 8.

^{page 798 note 1} Of the Mosque of Seville there remain only the western court and the minaret (the celebrated “Giralda”). The rest of the original site is occupied by the Cathedral, which covers the area of the mosque. The court which remains is the identical part of the mosque inaccurately referred to by Ibn Sāid, as “the garden in its midst” the fact being that the plan of this mosque, like that of Cordova, and apparently most West African mosques of the colonnaded type, differed from the Egyptian plan in having one huge forest of columns on the Kibla side, while the Sahn formed a forecourt to the whole. The existing court at Seville contains a fountain supposed to he the original fountain of ablution, and is planted with orange trees, most of which are said to date from the sixteenth century. According to O'Shea's Guide to Spain and Portugal (ed. 1885, p. 115), “There were always many trees in it, especially palms and cypresses, many of which were destroyed in a hurricane in 1822.”

^{page 799 note 1} Description de l'Egypte, second edition, Paris, 1829, vol. 18, p. 464. Who this gentleman was, ana when, where, and on what authority he made the above statement, we are not told.

^{page 800 note 1} Called the Madrasa Nasirīya and the Madrasa Kamhīya. They were devoted to theology after the sects of Ash Shāfaī and Ibn Mālik respectively (Al-Makrīzy, ii. 364–4).

^{page 800 note 2} Korān, lv. 26, 27.