[80] Marcellinus, loc. cit.; Zonaras, xiii, 22; Nicephorus Cal., xiv, 1, confuses the work of the two men. The Anon. Patria (Banduri), p. 20, says that the two factions of the circus, each containing eight thousand men, were employed on the work. Beginning at either end, they met centrally at a gate hence called “of many men” (Polyandra). Mordtmann (op. cit., p. 28) wholly rejects this tale, as it does not fit in with some of his identifications. It would, however, be well suited to the P. Rhegii, where the existing inscriptions are found. Some local knowledge must be conceded to an author of the twelfth century, who probably lived on the spot. Wall-building was a duty of the factions.
[81] Dionysius caused the Syracusans to build the wall of Epipolae, of about the same length, in twenty days (Diod. Sic., xiv, 18). The Peloponnesians built a wall across the isthmus against Xerxes in a short time (Herodotus, viii, 71, etc.). There was much extemporary wall-building at Syracuse during the siege by Nicias (Thucydides, vi, 97, etc.). The wall of Crassus against Spartacus was nearly forty miles long (Plutarch, Crassus). Except the first, however, these were more or less temporary structures. Very substantial extempore walls are frequently mentioned by both Greek and Latin historians as having been erected during sieges, etc. See especially Caesar (i, 8) and Thucydides (iii, 21, Siege of Plataea).
[82] The earliest and most reliable source is the Notitia Dignitatis utriusque Imperii, etc., which dates from the time of Arcadius. To this work is prefixed a short description of Rome and CP., which enumerates the chief buildings, the number of streets, etc., in each division of those cities. Next we have the Aedificia of Procopius, the matter of which, however, does not come within the scope of the present chapter. A gap of six centuries now occurs, which can only be filled by allusions to be found in general and church historians, patristic literature, etc. We then come to a considerable work, the Anonymous, edited by A. Banduri (Venice, 1729), a medley of semi-historical and topographical information, often erroneous, ascribed to the twelfth century. A second edition of this work, introduced by the Byzantine fragment of Hesychius of Miletus, passes under the name of Geo. Codinus, who wrote about 1460. Here we draw the line between mediaeval and modern authors, and we have next the Topography of CP., by P. Gyllius, a Frenchman, who wrote on the spot about a century after the Turkish conquest. His Thracian Bosphorus, which preserves much of the lost Dionysius of Byzantium, is also valuable. Later still comes the monumental CP. Christiana of Ducange (Paris, 1680), a mine of research, by one of those almost mediaeval scholars, who spent their lives in a library. Of contemporary treatises, which are numerous and bulky, I will only mention the following, from which I have derived most assistance: J. Labarte, Le Palais Impériale de CP., Paris, 1871; A. G. Paspates, Βυζαντιναὶ Μελέται, CP., 1877, and Βυζαντινα Ανάκτορα, Athens, 1885; W. Mordtmann, Esquisses topographiques de CP., Lille, 1891. Among books intended less for the archaeologist than for popular perusal, the only one worthy of special mention is Constantinople, Lond., 1895, by E. A. Grosvenor, a fine work, admirably illustrated, but the author relies too implicitly on Paspates, and he has emasculated his book for literary purposes by omitting references to authorities. The book also contains several absurd mistakes, e.g., “The careful historian who ... wrote under the name of Anonymos,” etc., p. 313. To the above must now be added the important, Byzantine CP., the Walls, by Van Millingen, Lond., 1899, a sound and critical work. Another beautiful work has also been recently issued, viz., Beylié, L’Habitation byzantine, Grenoble, 1902. A wealth of authentic illustrations renders it extremely valuable for the study of the subject. This chapter was begun in 1896, and in the meantime scholars have not been idle. As the Bonn Codinus gives inter-textually all the passages of the anonymous Patria which differ, as well as an appendix of anonymous archaeological tracts, I shall in future, for the sake of brevity, refer to the whole as Codinus simply in that edition.
[83] That is the pierced dome elevated to a great height on pendentives. The splendid dome of the Pantheon dates, of course, from Hadrian, but the invention of the modern cupola may fairly be assigned to the Byzantines. The conception, however, had to be completed by raising it still higher on a tour de dome, the first example of which is St. Augustine’s, Rome (1483); see Agincourt, Hist. of Art, i, 67.
[84] Procopius, De Aedific., i, 3; Nicephorus Cal., xv, 25.
[85] Κυκλόβιον or στρογγύλον; Procopius, ibid., iv, 8. Theophanes, an. 6165, p. 541, etc. Possibly it looked like the tomb of Caecilia Metella or a Martello Tower and was the prototype of the castle shown on the old maps as the “Grand Turk’s Treasure-house,” built in 1458 by Mohammed II within his fortress of the Seven Towers; Map by Caedicius, CP., 1889; Ducas, p. 317; Laonicus, x, p. 529. Most likely, however, it was a wall uniting five towers in a round. The Cyclobion is attributed to Zeno, about 480; Byzantios, Κωνσταντινούπολις, i, 312; Grosvenor, op. cit., p. 596.
[86] Grosvenor calls the existing road the remains of Justinian’s “once well-paved triumphal way,” I have found no corroboration of this assertion. From Constant. Porph. (De Cer. Aul. Byz., i, 18, 96, etc.), I conclude there was no continuous road here for many centuries afterwards. Paspates (op. cit., p. 13) thinks the last passage alludes to it as πλακωτῆ, but this is evidently the highway to Rhegium, etc. (Procop., De Aedific., iv, 8).
[87] Cod. VIII, x, 10; Procopius, De Bel. Pers., i, 25; Cinnamus, ii, 14; Anthol. (Planudes), iv, 15, etc.
[88] This fount is still extant and accessible beneath the Greek church of Baloukli (Grosvenor, op. cit., p. 485, etc.).
[89] Gyllius (Dionys. Byz.), De Bosp. Thrac., ii, 2; De Topog. CP., iv, 5.