The number of harbours found, at one time or other, on the southern shore of the city formed one of the most striking features in the aspect of Byzantine Constantinople. This was not due to any natural facilities offered by that shore for the purpose. On the contrary, although the outline of the coast is very irregular, it presents no bay where ships may be moored for the convenience of commerce, or into which they can find refuge from storms. The waves, moreover, cast up great quantities of sand upon the beach. Hence, all the harbours on this side of the city were, to a great measure, artificial extensions of some indentation of the coast, and their construction and maintenance involved great labour and expense. They ranked, in fact, among the principal public works of the capital. But the interests of commerce with the regions around the Sea of Marmora and with the Mediterranean were so great, and the difficulty which vessels coming from those regions often found to make the Golden Horn, owing to the prevalence of north winds, was so serious as to outweigh all drawbacks or impediments, and secured for the accommodation of the shipping frequenting this side of the city no less than five harbours. These harbours were probably constructed in the following chronological order: the Harbour of Eleutherius, known also as the Harbour of Theodosius; the Harbour of the Emperor Julian, known also as the New Harbour, and as the Harbour of Sophia; the Harbour of Kaisarius, the same probably as the Neorion at the Heptascalon; the Harbour of the Bucoleon; and the Kontoscalion. We shall consider them in the order of their position on the shore, proceeding from east to west.

Harbour of the Bucoleon.

The Harbour of the Bucoleon was attached to the Great Palace^[989] (τὸ τοῦ παλατίου νεώριον ἑν τῷ Βουκολέοντι) for the convenience of the emperor, who in a city like Constantinople would have frequent occasion to move to and fro by water. Its name was derived from a marble group of a Lion and a Bull upon the harbour’s quay, the lion being represented with his left foot upon a horn of the bull, in the act of twisting his victim’s head round to get at the throat.^[990] The harbour, partly artificial, was protected by two jetties from the violence of the winds and waves;^[991] and, in keeping with its destination, displayed considerable architectural splendour. Its quay was paved with marble,^[992] and adorned with figures of lions, bulls, bears, and ostriches;^[993] a handsome flight of marble steps led to the water;^[994] and upon the adjoining city walls rose two Imperial villas, known as the Palace of the Bucoleon (τὰ παλάτια τοῦ Βουκολέοντος).^[995]

Strangely enough, the site of a harbour so prominent, and so fully described, has been a point concerning which students of the topography of the city have widely differed. Dr. Paspates^[996] placed the harbour at a distance of 104 feet to the south of Indjili Kiosk, consistently with his opinion that the ruins discovered behind that Kiosk marked the site of the Palace of the Bucoleon.^[997] With much learning and ingenuity, Labarte argues that the Harbour of the Bucoleon was in the recess of the shore at Ahour Kapoussi.^[998] Von Hammer wavered in his opinion, placing the harbour at one time at Tchatlady Kapou, and at another at Kadriga Limani.^[999] And yet to Von Hammer is due the discovery of the evidence that puts an end to all uncertainty on the subject, by showing us that the marble group of the Lion and the Bull, which gave the harbour its name, stood at Tchatlady Kapou.

The evidence on the subject is found in a report which Pietro Zen, Venetian envoy to the Turkish Court, sent to his Government in 1532, where he describes the monument at great length, as he saw it after it had been shaken by an earthquake. In quoting this description,^[1000] Von Hammer, however, not only fails to use it for the settlement of the question at issue, but also omits portions of the report which are of the utmost importance for determining the exact site of the famous group. Dr. Mordtmann, citing Von Hammer, has appreciated the significance of the passage referred to, and employs it more successfully, but with the same omissions.^[1001]

The original manuscript of the report is preserved in the Marciana Library, among the unpublished Archives of the Venetian Republic,^[1002] and the passage with which we are concerned reads to the following effect:

“At the gate at which animals are slaughtered (near the columns of the Hippodrome, on the road below), which in Turkish is named Chiachadi Capisso, which in the Frank language means ‘Gate of the Crack,’ outside the said water-gate, and beneath the three ancient windows which have a lion at either end (of the row); there, down beside the shore, on two columns, is a marble block upon which is a very large bull, much larger than life, attacked at the throat by a lion, which has mounted upon the back of the (bull’s) neck, and thrown him down, and strikes at a horn of the bull with great force. This lion is considerably larger than life, all cut out of one piece of stone of very fine quality. These animals used to stand with their heads turned towards Asia, but it seems that on that night (the night of the catastrophe) they turned themselves with their heads towards the city. When this was observed next morning, the whole population of the place ran together to the spot, full of amazement and stupefaction. And every one went about discoursing upon the significance of the event according to his own turn of mind; a comet also appearing for many nights.”

The original is as follows, the words in italics being omitted by Von Hammer: “Alla porta dove si amaza animali, acosto dile colone dilprodramo, da basso via, e in Turcho si chiama chiachadi capisso, e in francho vol dir para di crepido, fuora dila dita porta de marina, sotto quelle tre fenestre antiquissime che hanno uno lione per banda, li abasso alla marina, sopra due colone, e una lastra di marmoro sopra la qual e uno granmo tauro, maior bonamente che il vivo, acanatto de uno lione, el qual li e montato sopra la schena, et lo ho atterato, et da una brancha ad un corno dil tauro in un grandissimo atto; e questo leone assai maior del vivo e tutto di una piera de una bona vena ouer miner. Questi animali soleano esser con le teste voltate verso Anatolia, et par che quella medema notte i se voltasseno con le teste verso Conple., il che la matina veduto tutta questa terra li e concorsa et ha fatto stupir e stornir tutta quest terra; et ogni uno va discorendo secondo le passione dil animo suo, stante una cometa apparsa per molte notte, questa cosa per il preditto rispetto ho voluto significar.”^[1003]

Nothing can be more explicit or more decisive.

There is no room to doubt that the monument described by Zen was the group of the Lion and the Bull, described, before him, by Anna Comnena and Zonaras.^[1004] His description might be a translation of the account given of the group by those writers. Nor is there any uncertainty as to the locality where Zen saw the monument. He indicates the site with a redundancy which makes misunderstanding simply impossible, and for which he may be pardoned, since minute particularity seldom distinguishes the statements of authorities on the topography of the city. According to the Venetian envoy, the monument stood on the quay outside the water-gate named Tchatlady Kapou, which was a gate below the Hippodrome, and near a slaughter-house. The group stood, he adds, beneath a row of three windows, adorned with a lion at either end, belonging to a very ancient building.

Now, the gate to which the name Tchatlady pertains is a matter of public notoriety, and every particular by which Zen marks the entrance he had in mind holds good of that gate. It is near the Hippodrome, and on the level ground below the race-course. On the western headland of the little bay in front of it, is an old slaughter-house, by which Leunclavius, likewise, identifies the gate Tchatlady Kapou, and from which he derived the name of the entrance;^[1005] while to the east of the gate stood, until recent times, a Byzantine palace, in the façade of which was a row of three windows, supported at either end by the figure of a lion. The palace is thus described by Leunclavius: “This gate (Tchatlady Kapou) has on one side of it the marble-framed windows of an ancient building or palace, which rests upon the city walls themselves.”^[1006] Gyllius refers to it in the following terms: “Below the Hippodrome towards the south is the Gate of the Marble Lion, which stands without the city among the ruins of the Palace of Leo Marcellus. The windows of the palace are of ancient workmanship, and are in the city wall.”^[1007] Choiseul-Gouffier^[1008] gives a view of the palace as seen in his day, and so does Canon Curtis, in his Broken Bits of Byzantium. The façade was torn down in 1871, and the lions have been placed at the foot of the steps leading to the Imperial School of Art, within the Seraglio enclosure.^[1009]

With this evidence as regards the site of the group of the Lion and the Bull, it is impossible to doubt that the Harbour of the Bucoleon was in the little bay before Tchatlady Kapou. And with this conclusion every statement made by Byzantine writers regarding the harbour will be found to agree.

That the shore of this bay was, like the Harbour of the Bucoleon, once richly adorned with monumental buildings, is manifest from the beautiful pieces of sculptured marble found upon its beach and in the water. Furthermore, the bay stands, as the Harbour of the Bucoleon stood, within easy reach of the site of the Great Palace. Here also are found the ruins of two Imperial villas, situated in the very position ascribed to the Palaces of the Bucoleon; namely, upon the city walls, at the waters edge, and one of them on a lower level than the other.^[1011] Such correspondence goes to make the site of the Harbour of the Bucoleon one of the best authenticated localities in the topography of Byzantine Constantinople.

Here, however, a question arises. How far is this conclusion, regarding the site of the Harbour of the Bucoleon, compatible with the received opinion that the palace on the bay before Tchatlady Kapou was the Palace of Hormisdas, the residence of Justinian the Great while heir-apparent;^[1012] and that the bay itself was the Harbour of Hormisdas (ὁ λιμὴν τὰ Ὁρμίσδου)?^[1013]

In the face of all the evidence we have that the Harbour and the Palace of the Bucoleon were in the bay to the east of Tchatlady Kapou, there is but one answer to the question. We must either abandon the view that the Harbour and the Palace of Hormisdas had anything to do with that bay, and maintain that they stood elsewhere, or we must conclude that they were the Harbour and the Palace of the Bucoleon, under an earlier designation.

Two considerations may be urged in favour of the former alternative. First, the Anonymus distinguishes between the two palaces in a way which seems to imply that they were different buildings. “The Palace of the Bucoleon,” he says, “which stands upon the fortifications, was erected by Theodosius the Younger;”^[1014] while of the Palace of Hormisdas he remarks: “The very large buildings near St. Sergius were the residence of Justinian when a patrician.”^[1015]

In the second place, the Anonymus^[1016] identifies the Harbour of Hormisdas with that of Julian. “What is called τὰ τοῦ Ὁρμίσδου,” observes the former writer, “was a small harbour where Justinian the Great built a monastery and called it Sergius and Bacchus, and another church, that of the Holy Apostles (SS. Peter and Paul), after receiving unction at the foot of the seats (of the Hippodrome), because of the massacre in the Hippodrome. It was named the Harbour of Julian, from its constructor.” Codinus^[1017] also identifies the two harbours, and adds, that the Harbour of Julian had served for the accommodation of ships before the Harbour of the Sophiôn was constructed; that it had long been filled up; and that Justinian the Great had lived there before his accession to the throne. But if on the ground of these statements we identify the Harbour of Hormisdas with that of Julian, as Banduri^[1018] and Labarte^[1019] maintain, then the Harbour of Hormisdas was not situated in the bay to the east of Tchatlady Kapou, but at Kadriga Limani, the undoubted site of the Harbour of Julian, to the west of the gate.^[1020] The Palace of Hormisdas, also, must then have been in that direction.

In the light, however, of all our knowledge on the subject, the identity of the two harbours just named cannot be maintained. John of Antioch,^[1021] a far more reliable authority than the Anonymus or Codinus, makes it perfectly clear that the Harbour of Julian (which he calls by its later name, the Harbour of Sophia) was different from any harbour in the quarter of Hormisdas. According to him, the troops collected by Phocas for the defence of the city against Heraclius occupied three positions—the Harbour of Kaisarius, the Harbour of Sophia, and the quarter of Hormisdas. At the first two points were placed the Greens, while the third position was held by the Blues. From this account of the matter it is evident that the Harbour of Julian was not the harbour in the quarter of Hormisdas. It is a corroboration of this conclusion to find that in the narrative of the same events, given in the Paschal Chronicle,^[1022] while no mention is made of the Harbour of Hormisdas, the Harbour of Julian is described as situated in another quarter, the quarter of Maurus (κατὰ τὰ λεγόμενα Μαύρου).

In favour of the alternative that the Palace and Harbour of Hormisdas were the Palace and Harbour of the Bucoleon under another name, may be urged all that goes to show that the former stood where the evidence furnished by Pietro Zen has obliged us to place the latter. The bay and palace on the east of Tchatlady Kapou stand close to what was unquestionably the district of Hormisdas; for the Church of SS. Sergius and Bacchus (Kutchuk Aya Sophia), a short distance to the west of the gate, was in that district.^[1024] It would be strange if a palace and harbour so near that district were not those known by its name.

The palace at Tchatlady Kapou answers, moreover, to the description which Procopius gives of the Palace of Hormisdas, the residence of Justinian, as near SS. Sergius and the Great Palace.^[1025] Its position agrees also with the statement of John of Ephesus that the Palace of Hormisdas was below the great Imperial residence.^[1026] Again, the style of the capitals and other pieces of marble, which have fallen from the palace at Tchatlady Kapou into the water, resemble the sculptured work in the Church of SS. Sergius and Bacchus, erected by Justinian. And lastly, the palace at this point was regarded as the Palace of Justinian when Bondelmontius visited the city in 1422. “Beyond Condoscali (Koum Kapoussi),” says that traveller, as he proceeds eastward, along the Marmora shore of the city, “was the very large Palace of Justinian upon the city walls” (“Ultra fuit supra mœnia amplissimum Justiniani Palatium”).

All this being the case, it seems unavoidable to conclude that the Palace and Harbour of Hormisdas were the Palace and Harbour of the Bucoleon, under an earlier name. The circumstance that the palaces are distinguished by the Anonymus presents, after all, no serious difficulty, but the reverse; for, as a matter of fact, there are two palatial buildings on the bay east of Tchatlady Kapou, at a distance of some 110 yards from each other, and on different levels. One of the buildings, probably the lower, might be the Palace of Hormisdas; the other, on higher ground, and nearer the gate—may be the palace to which the Anonymus referred as the Bucoleon.

It is in keeping with this view of the subject to find that the terms “Palace of Hormisdas,” “Port of Hormisdas,” are not employed by Byzantine authors to designate an Imperial residence or harbour, after the name Bucoleon came into vogue.

The earliest writer who refers to the Harbour of the Bucoleon is the Emperor Constantine Porphyrogenitus,^[1027] in the tenth century. Later writers,^[1028] it is true, employ the name when speaking of events which occurred in the reign of Michael I., and in that of Theophilus, in the course of the ninth century. But whether these writers do so because the name was contemporary with the events narrated, or because, when the historians wrote, it was the more familiar appellation for the scene of those events, is uncertain. Should the former supposition be preferred, it was early in the ninth century that the term “Bucoleon” first appeared.

On the other hand, the last author who alludes to the Palace of Hormisdas is the historian Theophanes, who died in 818. The passage in which the allusion is found refers, indeed, to matters which transpired in the seventh century, viz. to the execution of a certain David, Chartophylax of (the Palace of) Hormisdas, in the reign of Phocas. But the historian could hardly have described an official position in terms not still familiar to his readers.^[1029]

Accordingly, the designation “Palace of Hormisdas” disappears about the time when the term “Bucoleon” appears, and this is consistent with the supposition that the two names denoted the same building at different periods of its history.^[1030]

The Palace of Hormisdas was so named in honour of the Persian Prince Hormisdas, who had been deprived of the succession to the throne of his country by a conspiracy of nobles, and confined in a tower; but who escaped from his prison through the ingenuity of his wife, and fled to New Rome for protection at the hands of Constantine the Great. The royal fugitive was received with the honour due to his rank, and this residence was assigned to him because near the emperor’s own palace.^[1031] Later, the residence was occupied, as already intimated, by Justinian while Crown Prince, with his consort Theodora; and after his accession to the throne, was by his orders, improved and annexed to the Great Palace.^[1032] It appears in the reign of Justin II. as the abode of Tiberius, upon his being appointed Cæsar.^[1033] Under ordinary circumstances, Tiberius should have occupied apartments in the Great Palace. But the Empress Sophia was bitterly jealous of his wife Ino, and forbade her to show herself at Court, on any pretext whatever. Obliged, consequently, to find a home elsewhere, the Cæsar selected the Palace of Hormisdas, because its proximity to the Great Palace would allow him to enjoy the society of his family, and attend to his official duties. But the jealousy of the empress was not to be allayed so readily. It followed Ino to the Palace of Hormisdas with such intensity that the ladies of the Court dared not visit her even there; and it compelled her at last to leave the capital and retire to Daphnusium.

As already stated, when Heraclius appeared with a fleet, in 610, before the city to put an end to the tyranny of Phocas, he found the quarter of Hormisdas defended by the Faction of the Blues.^[1034]

During the tenth century, the port and palace, then called Bucoleon, received special marks of Imperial favour. Constantine Porphyrogenitus, noted for his devotion to the Fine Arts, adorned the quay of the harbour with figures of animals, brought from various parts of the Empire.^[1035] Possibly, the group of the Lion and the Bull was placed there by him. He also attached a fishpond to the palace.

Later, Nicephorus Phocas added a villa, which he made his usual place of residence.^[1036] It was probably the building with the row of three windows, supported by a lion at either end. A still more important change was introduced by the same emperor. His austere character, and the heavy taxes he imposed for the maintenance of the army, made him exceedingly unpopular, notwithstanding his eminent services as the conqueror of the Saracens. So strong did the hostile feeling against him become, that, returning once from a visit to the Holy Spring of the Pegè, he was mobbed at the Forum of Constantine, and narrowly escaped being stoned to death before he could reach the palace.^[1037] Rumours of a plot to dethrone and kill him were also in circulation. He therefore decided to convert the Great Palace into a fortress, and to provision it with everything requisite to withstand a siege.^[1038]

Accordingly, he surrounded the grounds of the Imperial residence with a strong and lofty wall, which described a great arc from the neighbourhood of Ahour Kapoussi on the east to Tchatlady Kapou on the west, and thus cut off the palace from the rest of the city.^[1039] Luitprand,^[1040] who saw the wall soon after its erection, says of it: “The palace at Constantinople surpasses in beauty and strength any fortifications that I have ever seen.” Within this wall the Palace of Bucoleon was, of course, included.

Labarte^[1041] and Schlumberger^[1042] maintain, indeed, that Nicephorus surrounded the Palace of Bucoleon with special works of defence, and constituted it a citadel within the fortifications of the Great Palace. But Leo Diaconus, Cedrenus and Zonaras, our authorities on the subject, make no such statement.^[1043]

As might be expected, historical events of considerable importance transpired at the Port and the Palace of the Bucoleon.

Here, in 919, Romanus Lecapenus, admiral of the fleet, made the naval demonstration which compelled Constantine VII. Porphyrogenitus to accept him as a colleague, and to surrender the administration of affairs into his hands.^[1044]

It was here that the memorable conspiracy against Nicephorus Phocas was carried out, in 969, by John Zimisces, with the connivance of the Empress Theophano.^[1045] Under cover of the night, the conspirators embarked at Chalcedon, the residence of Zimisces at the time, and in the teeth of a strong north wind, and with snow falling heavily, crossed to the Bucoleon. A low whistle announced their arrival to their accomplices, who were watching on the terrace of the palace; and in response, a basket held fast by ropes was stealthily lowered and raised, again and again, until one by one all in the boat were lifted to the summit. The last to ascend was Zimisces himself. Then the traitors made for the apartment in which they expected to find the emperor. Nicephorus, who had received some intimation of the plot, was not in his usual chamber, and the conspirators, fearing they had been betrayed, were about to leap into the sea and make their escape, when a eunuch appeared and guided them to the room in which the doomed sovereign lay fast asleep on the floor, on a leopard’s skin, and covered with a scarlet woollen blanket. Not to spare their victim a single pang, they first awakened the slumberer, and then assailed him with their swords as he prayed, “Lord, have mercy upon me.” As if to add irony to the event, Nicephorus met his fate, it is said, on the very day on which the fortifications around the palace were completed. After this, guards were stationed, at night, on the quay of the Harbour of the Bucoleon, to warn off boats that approached the shore.^[1046]

From this point, Alexius Comnenus entered the Great Palace, after the deposition of Nicephorus Botoniates; leaving his young wife and her immediate relatives in the residence by the shore, while he himself, with the members of his own family, proceeded to the higher palace (τὸ ὑπερκείμενον παλάτιον).^[1047] Here, also, in 1170, Amaury, King of Jerusalem, landed on the occasion of his visit to Manuel Comnenus, to seek the emperor’s aid against Saladin. Access to the palace by this landing, says William of Tyre,^[1048] in his account of that visit, was reserved, as a rule, for the emperor exclusively. But it was granted to Amaury as a special honour, and here he was welcomed by the great officers of the palace, and then conducted through galleries and halls of wonderful variety of style, to the palace on an eminence, where Manuel and the great dignitaries of State awaited the arrival of the king.

In the course of time, as the prominent position of the Palace and the Harbour of Bucoleon rendered natural, the name Bucoleon, it would appear, was extended to the whole collection of buildings which formed the Great Palace, facing the Sea of Marmora. That is certainly the sense in which Ville-Hardouin employs the term in his work on the Conquest of Constantinople by the Crusaders. He associates “le palais de Bouchelyon” with the Palace of Blachernæ, as one of the principal residences of the Greek emperors. In the division of the spoils of the city, the Palace of “Bouchelyon,” like the Palace of Blachernæ, was to belong to the prince whom the Crusaders would elect Emperor of Constantinople;^[1049] upon the capture of the city, the Marquis of Montferrat hastened to seize the Palace of Bucoleon, while Henry, the brother of Baldwin, secured the surrender of the Palace of Blachernæ;^[1050] the treasure found in the former is described as equal to that in the latter: “Il n’en faut pas parler; car il y en avait tant que c’était sans fin ni mesure.” Indeed, the statements of Ville-Hardouin concerning the Palace of Bucoleon make the impression that of the two Imperial residences which he names, it was, if anything, the more important.^[1051] Thither Murtzuphlus fled when his troops were discomfited.^[1052] There, the Marquis of Montferrat found congregated for safety most of the great ladies of the Court, including Agnes of France, wife of Alexius II., and Margaret of Hungary, wife of Isaac Angelus.^[1053] And to the Palace of Bucoleon, the richest in the world (“el riche palais de Bochelyon, qui onques plus riches ne fu veuz”), the Latin Emperor Baldwin proceeded in great state, after his coronation in St. Sophia, to celebrate the festivities attending his accession to the throne.^[1054] There, also, were held the festivities in honour of the marriage of the Emperor Henry with Agnes, the daughter of the Marquis of Montferrat.^[1055] It is not possible that the two comparatively small buildings at Tchatlady Kapou could be the palace which Ville-Hardouin had in mind in connection with these events. The terms he employs, in speaking on the subject, were appropriate only to the Great Palace as a whole.

The designation of the Palace of Bucoleon as “Chastel de Bouchelyon”^[1056] is no evidence that Ville-Hardouin used the name in its restricted sense, as Labarte contends. For the Great Palace was within a fortified enclosure, and could therefore be styled a castle with perfect propriety, just as the same historian, for a similar reason, speaks of the Palace of Blachernæ as a “chastel.” Nor does the fact that the Marquis of Montferrat reached the Palace of Bucoleon by riding along the shore (“chevaucha tout le long du rivage, droit vers Bouchelion”)^[1057] prove that the residence beside Tchatlady Kapou was the one he wished specially to secure. For the grounds of the Great Palace were thus accessible by a gate which stood at the eastern extremity of the Tzycanisterion, on the plain beside the Sea of Marmora, and which communicated with the quarter of the city near the head of the promontory.

Two incidents in Byzantine history, cited by Labarte^[1058] himself, establish the existence of such a gate, beyond contradiction. When Stephen and Constantine, the sons of the Emperor Romanus Lecapenus, deposed their father, in 944, and sent him to a monastery on the island of Proti,^[1059] great fears were entertained in the city, that a similar, if not a worse, fate had befallen his associate upon the throne, the popular Constantine VII., Porphyrogenitus. The people, therefore, crowded about the palace to ascertain the truth, and were reassured that their favourite was safe by his appearance, with dishevelled hair, at the iron bars of the gate which stood at the end of the Tzycanisterion (“Ex ea parte qua Zucanistrii magnitudo portenditur, Constantinus crines solutus per cancellos caput exposuit.”) The existence of a gate at this point is, if possible, still clearer from the statement of Constantine Porphyrogenitus,^[1060] that the Saracen ambassadors, after their audience of the emperor, left the palace grounds by descending to the Tzycanisterion, and mounting horse there. To approach the palace by that entrance evinced, therefore, no particular intention on the part of the Marquis of Montferrat to reach the buildings to which the name of Bucoleon strictly belonged. On the contrary, by that entrance one would reach the principal apartments of the Great Palace, sooner than the palaces beside the group of the Lion and the Bull, at Tchatlady Kapou.

The Bucoleon is mentioned for the last time in Byzantine history, in connection with the events of the final fall of the city. “To Peter Guliano, consul of the Catalans, was entrusted,” says Phrantzes,^[1061] “the defence of the quarter of the Bucoleon, and the districts as far as the neighbourhood of the Kontoscalion.”