¹ The substance of this article was included in a paper read at a meeting of the Archaeological Institute of America on 29th December, 1961, at the University of Michigan; a summary appeared in AJA 66 (1962), 197.

² e.g. Orpheus and the Argo: metope of the Sicyonian Treasury at Delphi; Fouilles de Delphes IV (plates), pl. 4. Also Orpheus and Eurydice with Hermes, relief in Naples: Brunn-Bruckmann, Denkmäler griechischer und römischer Skulptur, no. 341. For the Death of Orpheus in vase-painting, see Guthrie, W. K. C., Orpheus and Greek Religion, London, 1935, 64–5.Google Scholar

³ ‘La Mosaïque d'Orphée de Blanzy-les-Fismes (Aisne),’ Gallia XIII (1955), 41–77 (cited henceforth as Stern, Gallia).

^3a Segal, J. B., Archaeology 12 (1959), 157.Google Scholar This reference I owe to Mrs. S. Lloyd.

⁴ Fast. Arch., VII, 2310; Tekan, R., ‘Tarsus Mozaigi,’ IV Turk Tarih Kongresi, Ankara, 1952.Google Scholar

⁵ G. Fogolari, Fast. Arch., XIII, 3685.

⁶ Parlasca, K., Die römischen Mosaiken in Deutschland, 1959, pl. 2.Google Scholar

⁷ Foucher, L., Découvertes archéologiques à Thysdrus en 1960 (Tunis), 8–10, pls. 1–11.Google Scholar I am indebted to Mr. K. M. Phillips for this reference.

⁸ E. Vergara Caffarelli. Fast. Arch., VIII, 3887. The reference to an Orpheus mosaic at Beyrut, in Actes du V^e Congrès International d'Archéologie chrétienne (Vatican and Paris, 1957), 170Google Scholar, is erroneous; the figure surrounded by animals is the Good Shepherd; see Chéhab, M. H., ‘Mosaïques du Liban,’ Bulletin du Musée de Beyrouth XIV, Paris, 1957, 55 and pl. XXXI.Google Scholar

⁹ Stanton, G. R., ‘The Newton-St.-Loe Pavement,’ JRS XXVI (1936), 43–6Google Scholar; Toynbee, J. M. C., Art in Roman Britain (Soc. Prom. Rom. Studies, 1962), 14Google Scholar, nos. 185, 186, 195; Stern, Gallia, 75–6.

¹⁰ Bagatti, P. B., ‘Il mosaico dell'Orfeo a Gerusalemme,’ Riv. arch. crist. XXVIII (1952), 145–160Google Scholar; Stern, Gallia, 74–5.

¹¹ Sichtermann, H., ‘Archäologische Funde und Forschungen in der Kyrenaika 1942–1958,’ Archäologischer Anzeiger, 1959, coll. 239–348.Google Scholar Professor C. H. Kraeling's full report on the excavations of the Oriental Institute forthcoming.

¹² This project was carried out by the Cyrenaican Department of Antiquities under the writer's direction; the efficient assistance of Abdussalam Bazama, site-supervisor at Ptolemais, is gratefully acknowledged.

¹³ J. Lassus in his preface to Marec, E., Monuments Chrétiens d'Hippone (Paris, 1958), 5.Google Scholar The anticipated street was the eastern cardo.

¹⁴ A small bearded head (height from chin to crown 0.095 m.) of a fine-grain white marble, apparently of late-Antonine date, and four bronze coins, ranging from Hadrian to Anastasius I, were found in the fill but could not be associated with any stratification.

¹⁵ For the nimbus on Seasons, see Levi, Doro, Antioch Mosaic Pavements (Princeton, 1947), I, 289.Google Scholar Both here and in the Orpheus mosaic the average number of tesserae per square decimetre varies from 120 in the figure to 50 in the border.

¹⁶ In the Sculpture Museum at Cyrene, a second century mosaic depicting Dionysus has a close parallel for this black-spotted blue leopard.

¹⁷ cf. Ἀρχ. Ἐφ., 1917, 50, fig. 2, for a very similar bird at Nikopolis.

¹⁸ A Nilotic setting occurs frequently in Cyrenaican church-mosaics of the sixth century, notably at Gasr-el-Lebia and Cyrene: see Goodchild, R. G., ILN, 14th December, 1957, 1034–5Google Scholar; Ward Perkins, J. B., ‘A New Group of Sixth-century Mosaics from Cyrenaica,’ Riv. arch. crist., 1958, 183–192.Google Scholar

¹⁹ Doro Levi, o.c. (n. 15), 1, 359–363 (including a general discussion of Orpheus scenes and the paradeisos), and II, pls. CLXXV–CLXXVI. A local parallel would have been preferable, but see Ward Perkins, o.c., for the paucity of fifth-century mosaics in Cyrenaica. For the argument that these groups of wild and tame animals refer to the Messianic Peace predicted in Isaiah XI, see Stern, H., ‘The Orpheus in the Synagogue of Dura-Europos,’ Journal of the Warburg and Courtauld Institutes XXI (1958), 1–6.CrossRefGoogle Scholar

²⁰ Caputo, C. and Goodchild, R. G., ‘Diocletian's Price-edict at Ptolemais (Cyrenaica),’ JRS XLV (1955), 106–7Google Scholar; also Goodchild, R. G., ‘A Byzantine Palace at Apollonia (Cyrenaica),’ Antiquity XXXIV (1960), 246–7.CrossRefGoogle Scholar The palace, presumably the new gubernatorial seat, is dated by Goodchild ‘nearer 500 than 450’. The cutting of the aqueduct is known from Procopius, De Aed. VI, 2, where its restoration by Justinian is recorded. The increasing difficulties faced by Ptolemais early in the fifth century are vividly depicted in the correspondence of Synesius (Bishop c. 410). Probably at this period the city-wall of Ptolemais was dismantled and a smaller circuit built nearer the sea; see Goodchild, R. G., ‘The Roman and Byzantine Limes in Cyrenaica,’ JRS XLIII (1953), 74.Google Scholar Also, Goodchild in Quaderni di Archeologia della Libia 4, 83–95.

²¹ Guidi, G., ‘Orfeo, Liber Pater e Oceano in mosaici della Tripolitania,’ Africa Italiana VI (1935)Google Scholar; Stern, Gallia, 49–55.

²² This mosaic was badly damaged by torrential rains in October the same year. Two rectangular panels of equal width are surrounded, and separated, by a guilloche-border. The main panel with Orpheus and the animals is 1.50 m. wide and 1.80 m. high; the smaller panel below, a marine scene with fishes, is 0.66 m. high. At the feet of Orpheus is a small grey monkey confronting an elephant, behind which appear the head and shoulders of an antlered deer; higher up on the right a large tiger moving away turns its head to listen, as does the lion in the bottom lefthand corner; above the lion a goat is shown facing, and, higher still, a large dog (?), and a snake entwined around a small tree, which grows from the panel's border—the same motif as at Ptolemais. The average number of tesserae per square decimetre varies from no in the panels to 55 in the border.

²³ Delbrueck, R., Die Consulardiptychen (Berlin, 1925), pl. 62Google Scholar; Rumpf, A., Stilphasen der spätantiken Kunst (1955). pl. 16Google Scholar; Volbach, W. F., Frühchristliche Kunst (Munich, 1958), pl. 53.Google Scholar The missorium is securely dated to A.D. 388 by the superscription referring to Theodosius' decennalia.

²⁴ Grabar, A., L'Empereur dans l'Art Byzantin (Paris, 1936), 189 ff.Google Scholar

²⁵ The long chlamys is worn by Orpheus elsewhere, notably a century earlier at Lepcis Magna (Guidi, o.c. (n. 20), and Stern, Gallia, fig. 17), and Piazza Armerina (Gentili, G. V., La Villa Imperiale di Piazza Armerina (Rome, 1956), pl. 29)Google Scholar; its importance here lies in its conjunction with two innovations, the frontal posture and the nimbus.

²⁶ Weitzmann, K., Greek Mythology in Byzantine Art (Princeton, 1951), 68Google Scholar, discussing three medieval miniatures of Orpheus with the nimbus, states that ‘This Christianization is due to the influence of a picture of David’. (The miniatures are Vat. cod. gr. 1947. Fol. 149^v; Mt. Athos, Panteleimon. Cod. 6. Fol. 165^r; Paris, Bibl. Nat. Cod. Coislin 239. Fol. 122^v: Weitzmann, ibid., figs. 82–4). In the light of the nimbed Orpheus at Ptolemais, these medieval nimbi may perhaps be derived directly from a much earlier ‘Christianized’ Orpheus, the David scenes representing a collateral development. (I owe this reference to the kindness of Professor E. Kitzinger at Dumbarton Oaks; Professor Kitzinger was also good enough to show me the photographs at Dumbarton Oaks of the Orpheus mosaic at Salona, Stern, Gallia, 76.)

²⁷ It is not suggested that either the technician, who laid this mosaic, or the master of the house, who ordered it, intended on his own initiative any esoteric significance; but surely the wide acceptibility of this motif in Late Antiquity is partly to be explained by its susceptibility to a double interpretation.

²⁸ Wilpert, G., Le Pitture delle Catacombe Romane (Rome, 1903), 36 and 222 ff.Google Scholar

²⁹ Wilpert, ibid., pl. 37 (S. Callixtus); cf. also ibid., 224 (S. Priscilla).

³⁰ Chéhab, M. H., ‘Mosaïques du Liban,’ Bulletin du Musée de Beyrouth XIV (1957), pl. XXXI.Google Scholar

³¹ § 14.

³² Migne, PG LXVI. The Hymns were probably written towards the end of Synesius' life; cf. Fitzgerald, A., The Letters of Synesius of Cyrene (London, 1926), 31 ff.Google Scholar