Though the Danube never passed through this channel, these caverns, which no European travellers have noticed, are deserving of attention. They are also mentioned by Hájí Khalífah in his account of the village of Injighiz, near the mountain of Chatáljah (Rumelí und Bosna, p. 17); and may be easily visited, as they are not much out of the way in going from Adrianople to Constantinople. _Note 3, p. 17._—_Altí Mermer._ In the present day nothing is seen on the spot of Altí Mermer except the mosque of that name. Some of these columns, which were probably used to ornament it, may perhaps be seen in the interior. _Note 4, p. 23._—_Sieges of Constantinople._ It is here necessary to rectify some of the author’s mistakes by the more correct chronology of Hájí Khalífah and the Byzantines. Evliyá states that the first siege took place in the year 34 of the Hijreh: this, however, is probably only a mistake of the copyist. He confounds the second siege, which took place in the year 47 (A.D. 667). _Vide_ Theophanes and Cedrinus, who call the Arab general Yezid, (Ἵζεδ), with the third in 53 (A.D. 672), and in which Ayyúb was killed. No mention is made either by Hájí Khalífah or the Byzantine historians of the third siege. Theophanes merely records the siege of Tyane in the year 91 (A.D. 710). The fourth also, in 97, seems to refer to the fifth, which by Hájí Khalífah and Theophanes is recorded as having happened two years later, _i.e._ 99, in the first year of the reign of Leo I., the Isaurian, when the Arabs are said to have built the mosque of Galata, which bears their name, and that called the Gul-jámi (rose mosque) in Constantinople. This tradition seems to be derived from the ancient names of the churches; that at Galata having been built by one Areobinthus, which to the Turks sounded like _Arab_; and the Gul-jámi having been called the rose-church because it was formerly a house belonging to a person of the name of Triantaphyllus (a rose), and was afterwards converted into a church by Romanus Argyropulos in the year 1031: _vide_ Cedrinus. Evliyá takes no notice of the siege by the Bulgarians, under their chief Paganus, in the year 764. Bullardus erroneously reckons this the fifth siege, it being in fact the sixth after the five preceding ones by the Arabs; and the eighth, if the two sieges of the ancient Byzantium are reckoned. The sixth and seventh sieges are also erroneously stated by Evliya. The former of these, which he states to have been in the year 160 of the Hijreh, ought to be four years later, _viz._ 164 (A.D. 780), as it is evidently the same as that of Hárún-ur-rashíd, which took place then, and not, as Evliyá gives it, in the year 255, which is too late by a century, as is also his seventh siege.