[308] In 1203 the Crusaders and Venetians had forced the boom tower on the Galata side and loosed the chain; but it was then outside the city walls. In the time of Cantacuzenus, Galata had been enlarged so that the end of the chain was quite safe unless Galata were taken. The walls terminated, as may still be seen by the remaining towers, near Tophana.
[309] Leonard, and Sauli’s Colonia dei Genovesi in Galata, p. 158. Other similar instances are cited by contemporaries, but it is not necessary to suppose that Mahomet had ever heard either of the fable of Caesar’s attack upon Antony and Cleopatra or of a like feat performed by Xerxes. The Avars had made a crossing similar to that contemplated by Mahomet. The transport of the imperial fleet into Lake Ascanius in order to take possession of Nicaea in 1097 might possibly have been known to him.
[310] Λοιπὸν ὁ ἀμερᾶς τὰς τριήρεις φέρας ἐν μίᾳ νυκτί, ἐν τῷ λιμένι τῷ πρωῒ ηὑρέθησαν: Phrantzes, 251.
[311] Dethier places them on a small plateau now occupied by the English Memorial Church. [Note on Pusculus, book iv. line 482. Professor van Millingen (p. 231), in discussing the question of the position of St. Theodore, suggests that the sultan’s battery stood nearer the Bosporus than the present Italian Hospital. This suggestion is not necessarily at variance with the position indicated by Dethier.]
[312] Philelphus, book ii. line 976: ‘Genuae tunc clara juventus obstupuit.’ Ducas, however, states that the Genoese claimed to have known of the proposed transport and to have allowed it out of friendship to Mahomet.
[313] ‘Et hic quidem in superiori parte per montem navigia transportavit ... in litore stabant milites parati propulsare hostes bombardis, si accederent prohibituri deducere naves.’ Chalcondylas, book viii.
[314] Crit. says 68; Barbaro, 72; Tetaldi, between 70 and 80; Chalcondylas, 70; and Ducas, 80; Heirullah says there were only 20; the Janissary Michael, 30; the Anon. Expugnatio, edited by Thyselius, sect. 12, says not less than 80.
[315] ‘Lacertus’ is the word Leonard ingeniously uses for the Greek πῆχυς.
[316] Crit. book iv. ch. 42. It is difficult to determine the size of the boat selected for this overland transit. Barbaro says, ‘le qual fusti si iera de banchi quindexe fina banchi vinti et anchi vintido’ (page 28). This would agree fairly well with the statement of Chalcondylas, that some had thirty and some fifty oars. Mr. Cecil Torr calculates that a thirty-oared ship would be about seventy feet long, a statement which appears probable (Ancient Ships, p. 21). The mediaeval galleys and other large vessels propelled by oars differed essentially from those of the sixteenth century, which were worked with long oars. See note on p. [234]. I am myself not entirely satisfied that among the boats were not biremes and possibly triremes in the sense of boats which had two or three tiers of oars, one above the other. Fashions change slowly in Turkey, and I have seen a bireme with two such tiers of oars on the Bosporus. No writer mentions the length of the vessels which were carried across Pera Hill. A large modern fishing caique in the Marmora, probably not differing much in shape from the fustae then transported, and containing twelve oars, measures about fifty feet long. When the boats are longer, two men take one oar, but this is very unusual. Leonard speaks of the seventy vessels as biremes. Barbaro calls them fustae. The former was probably the best Latin word to signify the new form of vessel. Many of the ships were large, though it may be taken as certain that none were of the length of the two galleys recently raised in lake Nemi, near Rome, which belonged to Caligula, each of which is 225 feet long and 60 feet beam.
[317] See note in Appendix on transport of Mahomet’s ships.