University of Oxford

File(s) not publicly available

E00591: Mosaic in the Rotunda of Thessalonike (south Balkans/Greece), of the 5th/6th c., depicting *Onesiphoros and Porphyrios (soldier and servant, martyrs of Parion, S00467), with inscriptions describing Onesiphoros as a soldier, and indicating August as the month of their festival.

online resource
posted on 2015-06-09, 00:00 authored by erizos
The two figures are depicted on the south panel of the mosaic circle, and can be with confidence identified as the follower of Paul the Apostle, Onesiphoros of Ikonion and his servant Porphyrios, both martyred at Parion in Hellespontus.

Both figures are depicted as youthful men, clean shaven and with rich curly hair, in the ‘mop style’ fashionable in the late 5th and 6th centuries. Onesiphoros wears a white chlamys with purple/blue tablium over a white tunic with blue segmentum and a gold girdle (cingulum). His figure is accompanied by the inscription:

στρ(ατιώτου) μηνὶ

‘(Memory) of Onesiphoros, soldier, in the month of August’

Onesiphoros’ servant Porphyrios is depicted as a commoner, wearing a purple paenula over a yellow/greenish tunic, and is accompanied by the inscription:

μηνὶ Αυ-

‘(Memory) of Porphyrios in the month of August’

Summary and translation: Efthymios Rizos.


Evidence ID


Saint Name

Onēsiphoros and Porphyrios, martyrs in Parion : S00467

Saint Name in Source

Ὀνησίφορος, Πορφοίριος

Type of Evidence

Images and objects - Wall paintings and mosaics Inscriptions - Formal inscriptions (stone, mosaic, etc.)


  • Greek

Evidence not before


Evidence not after


Activity not before


Activity not after


Place of Evidence - Region

Balkans including Greece

Place of Evidence - City, village, etc


Place of evidence - City name in other Language(s)

Thessalonike Drizypera Δριζύπερα Drizypera Büyük Karıştıran

Cult activities - Festivals

  • Saint’s feast

Cult activities - Places

Cult building - independent (church)

Cult activities - Use of Images

  • Public display of an image


The circular building known as the Rotunda of Saint George was built in the early 4th century as a part of the palatial complex of Thessalonike, probably under the emperor Galerius (Caesar to Diocletian 293-305, Augustus 305-311). Originally built as an imperial mausoleum or pagan temple, the building was later consecrated as a Christian church and functioned as such continuously until the 16th century, when it was converted into a mosque. As a church (the dedication of which is unknown), the Rotunda probably preserved a special association with the palace which was functional at least until the 7th century. The conversion of the building into a church was accompanied by the decoration of its dome with sumptuous mosaics which survive in a very fragmentary state. Various chronologies have been suggested, ranging from the late 4th to the 6th centuries, the likeliest period probably being 450-550. This is based on iconographic evidence (hairstyle and costume of the figures) and on the results of chemical analysis of the mortars of the mosaic (see the discussion by Fourlas 2012, 177 ff.). The mosaics present no signs of secondary phases or repairs. Their extensive damage seems to be due to natural causes. The mosaics consisted of a visionary scene (theophany) in three concentric registers. The figure of Christ in glory is depicted in a mandorla in the centre (now mostly lost), carried by four angels. This is followed by a register of saints (apostles, prophets, martyrs, or angels) in procession on a green ground – perhaps depicted according to the apocalyptic theme of the 24 Elders. Only the feet of some of these figures can be seen, and their precise number is unknown. The only part of the mosaic surviving in good condition is the third and lowest register, which is divided into eight panels (seven of which are preserved) with figures of praying martyrs (19 or 20 in number, of whom 16 are preserved). They are depicted standing with their hands raised in prayer, in front of sumptuous architectural façades, and are accompanied by inscriptions indicating their name, often their profession, and the month of their festival. The figures of the martyrs are depicted in official late antique dress, by which they are distinguished as soldiers (wearing the military cloak, the chlamys) and civilians (wearing the civilian cloak known as a paenula). Their dress also distinguishes these martyrs from the figures of the upper register, which were depicted wearing tunic and pallium/himation (the philosopher’s habit). Thus a sense of hierarchy is created, which was also accentuated by the greater size of the lost figures of the upper register. The mosaics depict the physiognomic characteristics of the martyrs with realism and precision, expressing differences in age and social status. Laymen and soldiers are depicted wearing their hair in the mop style, fashionable in the late 5th and 6th centuries, and with their faces clean shaven or with stubble beard. By contrast, Christian clerics and scholars (the two physicians) have short hair and longer beards. Although some scholars have proposed interpretations of the figures of the lower register as portraits of non-saintly persons, perhaps of donors, most of them can safely or tentatively be identified with martyrs known from the hagiographical sources. None of them seems to be a local saint of the city of Thessalonike, but they represent various parts of the East Roman Empire, including Anatolia, the Balkans, the Levant, and Egypt. No saints from the West can be identified. The 10th century Synaxarium of the Church of Constantinople suggests that many of the saints shown were honoured with chapels and feasts in the capital. The reasons and criteria for the selection of these particular saints in the dome of the Rotunda are unknown. Their placement does not follow any apparent logic. It is possible that the church possessed relics of these martyrs (see Kleinbauer 1972, 55), or that these figures were particularly revered by aristocratic families related to the palace of Thessalonike and its church. According to Ernst Weigand, the mosaic cycle of the Rotunda represents a selective calendar of saints from around the Christian world and the ecclesiastical year (Weigand 1939). The accompanying inscriptions are of major interest, since they emulate the formulation known from the earliest hagiographic calendars (martyrologia) (on them, see Feissel 1983). These inscriptions very probably echo the calendar observed in the church of Thessalonike, and possibly in the whole of Illyricum, of which Thessalonike was the ecclesiastical capital. It is remarkable that the feast dates present more similarities with the ecclesiastical calendar of Rome, than with that of Constantinople (as known in its post-Iconoclasm form, reflected in the Synaxarium of the Church of Constantinople).


The figures of Onēsiphoros and Porphyrios are one of two pairs of martyrs depicted in the mosaics of the Rotunda (the other being *Kosmas and Damianos, on whom see E00590). This is the only depiction of the two martyrs known from Late Antiquity, while the accompanying inscriptions are the only source for them having a festival in August. Identified as a companion of Paul the Apostle (mentioned 2 Tim. 1:16-18 and 4:19), Onēsiphoros was revered as a martyr alongside his servant and companion Porphyrios. Their hagiography describes Onēsiphoros as a man of noble birth from Ikonion in Asia Minor. The 10th century Synaxarium of Constantinople preserves the strange piece of information that he was of the gens Anicia (ἐκ γένους τῶν Ἀννικιόρων; Acta Sanctorum Novembris, Propylaeum, p. 823). This may explain why he is depicted and described as a soldier in our mosaic: Onēsiphoros was imagined as an official in the service of the emperor. Interestingly, the profession of his servant, Porphyrios, is not indicated by his inscription. The fact that August is given as the feast month of both figures suggests that they were already celebrated as a pair. No other source preserves the memory of this festival. The Synaxarium of Constantinople dedicates to them a quite long entry on 16 July (Acta Sanctorum Novembris, Propylaeum, p. 823-4). On the inscription, see: Feissel 1983, 105.


Bakirtzis, C., “Rotunda,” in: C. Bakirtzis (ed.), Mosaics of Thessaloniki : 4th to 14th Century (Athens: Kapon, 2012), 51-117 (with the most recent and best pictures). Feissel, D., Recueil des inscriptions chrétiennes de Macédoine du IIIe au VIe siècle (Bulletin de correspondance hellénique. Supplément; Athènes, Paris: Ecole française d'Athènes, De Boccard, 1983), 103-110 (for the inscriptions). Fourlas, B., Die Mosaiken der Acheiropoietos-Basilika in Thessaloniki : eine vergleichende Analyse dekorativer Mosaiken des 5. und 6. Jahrhunderts (Millennium-Studien; Berlin, Boston: De Gruyter, 2012), 156-195 (with a full survey of bibliography). Kleinbauer, W.E., “The Iconography and the Date of the Mosaics of the Rotunda of Hagios Georgios, Thessaloniki,” Viator 3 (1972), 27-107 (extensive discussion of the hagiographic questions). Spieser, J.M., Thessalonique et ses monuments du IVe au VIe siècle : contribution a l'étude d'une ville paléochrétienne (Bibliothèque des écoles françaises d'Athènes et de Rome; Paris: Boccard, 1984), 125-164. Torp, H., La rotonde palatine à Thessalonique : architecture et mosaïques, 2 vols., Athènes: Kapon, 2018. (with the most recent and best pictures) Weigand, E., “Der Kalenderfries von Hagios Georgios in Thessalonike. Datierung, Ideen- und kunstgeschichtliche Stellung,” Byzantinische Zeitschrift 39 (1939), 116-145.

Usage metrics

    Evidence -  The Cult of Saints in Late Antiquity



    Ref. manager