^{page 105 note 2} Salmon's ed., 59-61; transl., 145-7; translated into German by Herzfeld, op. cit. 11. 135-7. Epitome in Streck, op. cit., 63-4. See also Strange, Le, Baghdad, 33–7Google Scholar. Mr. Rhuvon Guest has kindly revised the translation given here.

^{page 105 note 3} Herzfeld (op. cit. II. 135, n. 5) remarks: ‘That is the exact meaning of the word; ‘aqab expresses the idea of the ends of the two pieces. One must likewise understand two pieces of column, one above the other. In Sāmarrā we have a complete analogy in the binding of the three lengths of marble column at [each of] the four corners of a composite pier. The junction is effected by means of mortising and, since the necessary strength is not imparted by that alone, by steel rings and cement also. The word for the kind of binding (mu'aqqab)….’

^{page 105 note 4} Lane gives ‘bound with ‘aqab, a sinew or tendon of which bow-strings are made’.

^{page 105 note 5} Streck (op. cit., 63) omits this obscure passage. Salmon (145-6) translates it as ‘… se composaient de deux morceaux aboutés l'un à l'autre au moyen de cordes (?) de colle et de crampons de fer, à l'exception de cinq ou six colonnes près du minaret, car dans chacune de celles-ci, il y avait des morceaux ajustés tout autour du bois de la colonne’. Le Strange (Baghdad, 34) paraphrases it as: ‘Most of these columns were constructed of two or more beams or baulks of timber, joined together endwise with glue, and clamped with iron bolts; but some five or six columns, those near the minaret, were each of a single tree-trunk. All the columns supported round capitals, each made of a block of wood, which was set on the shaft like a drum.’ Herzfeld (op. cit. II. 135, n. 6) remarks: ‘It is certain that the difference in the columns excluded consisted in the fact that they, in contrast to the others, were composed of one piece of wood. With the sequel fī kull ustuwāna qiṭa'an mulaffaqa muḍauwara min ḫashab al-asāṭīn begins a new train of ideas referring to the columns generally. The word mulaffaqa which I translate as capital, Salmon as ajustés, and from which Le Strange has obviously obtained his capitals set like a drum, is difficult to understand. The meaning must be ‘capital’: laffaqa I and II according to Freytag: junctis duabus parti-bus consuit (pannum), Qāmūs and the Dīwān of Hudhail; according to Dozy: II, consuo; from Jubayr, 68, I; a kind of ship in the Red Sea, in Idrīs¯, Klim., II, sect. 6: al-marākib al-mulaffaqa contrasted with al-maräkib min qiṭa‘ wāḫida; agencer, ajuster. Accordingly these capitals must have been made of several pieces fastened together, like modern wooden capitals in Mesopotamia and Persia. In Mōṣul people call the volutes of capitals (common technical term) dikma malfūf. I should like to be able to write mulaffafa and translate it by ‘turned’ or ‘provided with volutes’’.

^{page 106 note 1} Also Ṭabarī, III. 322; and Muqaddasī, 121, l. 7. Ibn Al-Athīr (V. 439, ll. 7-10) points out that this inaccuracy was due to the fact that the mosque was built against the side of the palace, after the completion of the latter, which had no special orientation.

^{page 106 note 2} Ṭabarī, III. 321; Yāqūt, Mu'jam, 1. 681, l.II; Ibn Al-Athīr, V. 439, l. 7.

^{page 106 note 3} Herzfeld remarks (136, n. 1): ‘Only the enclosing walls were renewed in brick and mortar, the wooden columns remained, or were replaced by similar ones, as follows from the express mention of the master-carpenter.’ This is not conclusive, for he must in any case have been needed for the roof; nevertheless, it is probable, as we shall see, that the new mosque, like the old, had wooden columns.

^{page 106 note 4} 109, ll. 3-4. He describes it as ‘built of kiln-baked bricks (ājurr) and gypsum (jiṣṣ); raised on columns of teak (asāṭīn as-sāj), and with a ceiling of teak-wood decorated with lapis lazuli’.

^{page 107 note 1} 250× 167m. = 41, 750 sq. m. against 103·6 × 103·6 = 10, 733 nearly. Herzfeld (II. 137) by a slip says ‘twenty-five times as great’.

^{page 107 note 2} 431, l. 7; and Herzfeld, op. cit. II. 137.

^{page 107 note 3} Op. cit. II. 137.

^{page 107 note 4} Op. cit. II. 137-8.

^{page 109 note 1} See my Early Muslim Architecture, I. 18–19 Google Scholar.

^{page 109 note 2} Strange, Le (Baghdad, 35)Google Scholar chooses the southeast, a view which is obviously untenable.

^{page 109 note 3} Lane says: ‘The upper, or highest, part (or end) of a sitting-room or sitting-place’, citing the Tāj al-'Arūs.

^{page 111 note 1} 109, l. 4.

^{page 111 note 2} Asher's transl., I. 96-7.

^{page 111 note 3} Wright's ed., 227 and 230; de Goeje's ed., Schiaparelli's transl., 214 and 217.

^{page 111 note 4} Le Strange, op. cit., 37.

^{page 111 note 5} 11.107.

^{page 111 note 6} Op. cit., 37.

^{page 111 note 7} Sinclair's transl., Hakluyt Socy., 64-5. It would also appear to be referred to in the expression ‘… vne grande Mosquée ruinée vis à vis de l'autre costé de la riuière’—in the Relation d'vn Voyage de Perse faict es années 1598 & 1599, par vn gentilhomme de la suite du Seigneur Seierley Ambassadeur du Roy d'Angleterre, on p. III of Lambert's, Caesar Relation (Paris, 1651)Google Scholar.

^{page 111 note 8} Gulshan-i-Ḫulafā (Constantinople, 1143 ff.), fol. 73 b, 11. 4-7, translated by Huart, , Histoire de Baghdad dans les temps modernes, 59 Google Scholar.

^{page 111 note 9} Voyage en Arabie (Amsterdam ed.) II. 239 ffGoogle Scholar.