¹ Tafrali, O., Topographie de Thessalonique (Paris, 1913)Google Scholar, plan at end.

² Makaronas, Ch., ὀκτάγονον τῆς θεσσαλονίκης, Praktika 1950, 303–321Google Scholar; idem, Χρονικὰ Ἀρχαιολογικά Makedonika ii (1954–52), 594–6.

³ Karamanoli-Siganidou, M., Ἀνασκαφαί Πλατείας Ναυαρίνου ArchDelt xx (1965), Chron., 408–9Google Scholar, Fig. 1, pls. 454, 456–8; Petsas, Ph., Ἀνακτορικὸν σύμπλεγμα Гαλερίου ArchDelt xxi (1966), Chron., 332Google Scholar, Fig. 1, 333, pl. 340 β and γ 341; idem, ArchDelt xxii (1967), Chron., 387–9, Figs. 7–9 (plans), 391 (plans reproduced at 1:100 in Petsas, , Χρονικά Ἀρχαιολογικά Makedonika viii [1969], pls. 1–2Google Scholar between pp. 152–3).

^3a cf. the dimensions of 100 X 200 Roman feet in the Basilica at Trier. Boethius, A. and Ward-Perkins, J. B., Etruscan and Roman Architecture (1970), 518.Google Scholar

⁴ Hébrard, E., ‘Les travaux du Service Archéologique de l'Armée d'Orient à l'arc de triomphe “de Galère” et à l'église Saint-Georges de Salonique’, BCH xliv (1920) 23, Fig. 9.Google Scholar

⁵ Makaronas, op. cit. 309, Fig. 6.

⁶ BCH lxxviii (1954), 138.

⁷ BCH lxxx (1956), 318.

⁸ Petsas, Ph., Гαλεριανὸν σ>γκρότημα Arch. Delt 24 (1969) Chron., 295–7Google Scholar, Fig. 3, pl. 305 β.

⁹ Vickers, M., ‘The Hippodrome at Thessaloniki’, JRS lxii (1972), 25–32.Google Scholar

¹⁰ e.g. (1) On the north side of Plateia Navarinou, Papadopoulou, Ph., Ἀνασκαφαί Πλατείας Ναυαρίνου θεσσαλονίκνς ArchDelt xix (1964), Chron., 332, pl. 378Google Scholar α and β Pelekanides, S., ‘Die Malerei der konstantinischen Zeit’, Akten des VII. Internationalen Kongresses für christliche Archäologie, Trier, 1965 (Vatican City/Berlin, 1969), 216–17Google Scholar, Figs. 1–3, pls. 111–12. (2) In the building discovered in 1968 on Odos Dimitriou Gounari, a plan of which is published in Petsas, Ph., Αίγαί–Πέλλα–θεσσλονίκή, Symposium ‘Ancient Macedonia,’ (Thessaloniki, 1969)Google Scholar, plan Γ opposite p. 244.

¹¹ Libanius, Antiochus 204–7 (= Or. xi, 6 ed. R. Foerster [Leipzig, 1903] 507–8): Cf. Downey, G., ‘The palace of Diocletian at Antioch’, Annales Archéologiques de Syrie iii (1953), 108Google Scholar (hereafter, Downey, 1953); idem, A History of Antioch in Syria (1961), 319 (hereafter, Downey, 1961).

¹² The clearest impression is given by Adam, R., Ruins of the Palace of the Emperor Diocletian at Spalato in Dalmatia (London, 1764), pl. 7.CrossRefGoogle Scholar cf. Hébrard, E. and Zeiller, J., Spalato, le palais de Dioclétian (Paris, 1912)Google Scholar, ‘Vue génerále à vol d'oiseau’ (frontispiece), and 48–52, plate: ‘Façade sur la mer. état actuel’ (between pp. 48 and 49).

¹³ Gose, E., ‘Die kaiserliche Sommerresidenz in Konz, Landkreis Saarburg’, Germania xxxix (1961), 204–6Google Scholar; Reusch, W. (ed.), Frühchristliche Zeugnisse im Einzugsgebiet von Rhein und Mosel (1965), 150–2Google Scholar; Wightman, E. M., Roman Trier and the Treviri (1970), 165–8.Google Scholar

¹⁴ Dyggve, E. and Vetters, H., ‘Mogorjelo; ein spätantiker Herrensitz im römischen Dalmatien’, Schriften der Balkankommission, Antiquarische Abtg. xiii (1966), 53–4Google Scholar, Figs. 24–5.

¹⁵ cf. the remarks of Duval, N. ‘La place de Split dans l'architecture antique du bas-empire’, Urbs iv (Split, 1961–1962), 93Google Scholar, and of Swoboda, K. M., Römische und romanische Paläste, 3rd edn. (1969), 133 ff.Google Scholar, 288, n. 50.

¹⁶ Downey 1953, Pl II, 1961, Fig. II.

¹⁷ Oelmann, F., ‘Zur Deutung des römischen Kerns im Trierer Dom’, Bonner Jahrbücher cxxvii (1922), 130–188Google Scholar; Wightman, op. cit., 98, 121, Fig. 12; Reusch, W. in Rettet das römische Trier, (Denkschrift der Archäologischen Trier-Kommission (1972)), 9–14Google Scholar, and plan.

¹⁸ cf. Eutropius, ix, 28: ‘Diocletianus privatus in villa, quae haud procul a Salonis est …’ St. Jerome, Chron. a. 2332: ‘ Diocletianus haud procul a Salonis in villa sua Spalato moritur.’ See Duval, op. cit. 70, for a detailed discussion of this question.

¹⁹ De mort, pers vii, 8.

²⁰ op. cit. (n. 2), 316.

²¹ Revue des Etudes Byzantines x (1952), 205.

²² ‘Die Datierung der Reliefs am Hadrianstempel in Ephesos und das Problem der tetrarchischen Skulptur des Ostens’, Istanbuler Mitteilungen xviii (1968), 248–9.

²³ Early Byzantine Churches in Macedonia and Southern Serbia (London, 1963), 123–4. Hoddinott would have been on slightly surer ground had he identified the plants as cypresses, for such trees were sacred to Sol in Syria; see the Palmyrene altar to Sol in the Capitoline Museum which has cypresses on one side: Stuart Jones, H., The Sculptures of the Museo Capitolino (Oxford, 1912), pl. 9Google Scholar, II, lc; Cumont, F., Syria ix (1928), 104–6Google Scholar, pl. 39, right; Buisson, Comte du Mesnil du, Les tessères et les monnaies de Palmyre (Paris, 1962), 116Google Scholar, Fig. 70 (pp. 264–5, 438–9 and 558–9 in particular deal with Sol and cypresses). A cypress is visible on a relief on the Arch of Galerius, next to what H. P. Laubscher identifies as a Temple of Sol (see his forthcoming Der Reliefschmuck des Galeriusbogens in Thessaloniki, under panel B II 17 [whence the references above] also Ch. Makaronas, Ἡ καμάρα, τὸ θριαμβικὸ τόξο τοῦ Γαλερίου στὴ Θεσσαλονίκη, Thessaloniki 1969, Fig. 29).

²⁴ This possibility was first suggested, only to be dismissed, by Hoddinott, I.c.

²⁵ e.g. Gellius, Aulus, Noct. Att. iii, 6Google Scholar: ‘Ideo in certaminibus palmam signum esse victoriae placuisse, quoniam ingenium eiusmodi ligni est, ut urgentibus opprimentibusque non cedat’ (quoting Plutarch); St. Gregory the Great, In Ez. hom, xvii: ‘Quid per palmam, nisi praemia victoriae, designatur?’ Cf. the punning tombstone from Rome with a palm branch and the inscription ΒΙΚΤΩΡΙΑ +, van der Meer, F. and Mohrmann, C., Atlas of the Early Christian World (1958), 141Google Scholar, Fig. 463.

²⁶ The relevant texts are conveniently assembled in Aufhauser, J. B. (ed.), Konstantins Kreuzesvision (Bonn. 1912).Google Scholar References to recent literature are to be found in Sordi, M. II, cristianesimo e Roma (1965), 452–3Google Scholar; and in Alföldi, A., The Conversion of Constantine and Pagan Rome (1948), viii.Google Scholar

²⁷ Guarducci, M., I graffiti sotto la confessione di S. Pietro ii (1958), 14 ff.Google Scholar; idem, La Tomba di Pietro (1959), 117–8; idem, The Tomb of St. Peter (1960), 123 ff. Accepted by Frazer, P., JRS lii (1962), 216–17Google Scholar, but not everyone would agree with Guarducci's reading, see reviews of I graffiti by, e.g., Ferrua, A., RACrist xxxv (1959), 245–6Google Scholar (Guarducci's reply, Archeologia Classica xiii [1961], 230–1, Fig. 44), and Toynbee, J.M.C., The Dublin Review, No. 481 (1959), 241.Google Scholar

²⁸ The symbol appears on medallions of Constantine dated to 315, Alföldi, A., ‘The initials of Christ on the helmet of Constantine.’ Studies in Roman Economic and Social History in honour of A.C. Johnson (1957), 303–311Google Scholar; Kraft, K., ‘Das Silbermedallion Konstantins mit dem Christusmonogram auf dem Helm.’ Jahrbuch für Numismatik und Geldgeschichte v–vi (1954–1955), 151–178Google Scholar, pls. 11–12; and in 317 on the obverses of a few coins of Siscia, again on Constantine's helmet, Roman Imperial Coinage vii (1966), 417–18. Cf. Toynbee's, J. M. C. observations in ‘A new Roman mosaic pavement found in Dorset’, JRS liv (1964), 10–11Google Scholar.

²⁹ Knipfing, J. R., ‘The Edict of Galerius reconsidered’, Revue Belge de Philologie i (1922), 693CrossRefGoogle Scholar; cf. A. Alföldi, op. cit. 9–10.

³⁰ Weigand, E., ‘Baalbek und Rom, die römische Reichskunst in ihrer Entwicklung und Differenzierung’, Jdl xxix (1914), 89Google Scholar; Heilmeyer, W. D., Korinthische Normalkapitelle, (RM ErgH xvi, 1970), 169–170.Google Scholar

³¹ Stuart, J. and Revett, N., The Antiquities of Athens iii (1794), pls. vi–viiiGoogle Scholar; Heilmeyer, op. cit. 74, pl. 18, 1–2; Travlos, J., Pictorial Dictionary of Ancient Athens (1971), 254–7Google Scholar, Figs. 325–8.

³² Weigand, op. cit. 90; Heilmeyer, op. cit. 74–7; see also E. von Mercklin, Antike Figuralkapitelle (1962) passim.

³³ Hoepfner, W., ‘Herakleia Pontike, Eine baugeschichtliche Untersuchung. Forschungen an der Nordküste Kleinasiens’, Osterr. Akad d. Wiss., Phil.-hist. Kl., Denkschr. lxxxix (1966).Google Scholar See too, the capitals from Ivajlovgrad, Bulgaria: Mladenova, J.. ‘La villa romaine d'Ivajlovgra’, Actes du premier congrès international des études balkaniques et sud-est européennes (Sofia, 1970), 527–34Google Scholar; ead., ‘Chapiteaux de revêtement de la villa près d'Ivajlovgrad’, Bulletin de l'Institut d'Archéologie xxxii (Belgrade, 1970), 129–147 (of second-century date).

³⁴ The right-hand volute is restored. Karamanoli-Siganidou, ArchDelt cit. (n. 3), 409, pl. 458β; Petsas, Makedonika vii (1967), 297, pl. 10; Brenk, op. cit. (n. 22), 248–250, pl. 74, 2; BCH xcii (1968), 904, Fig. 6. All the capitals are now in the Thessaloniki Archaeological Museum, and, when complete, they are all approximately 0.61 m high, and 0.82 m wide above, and 0.61 m below.

³⁵ Karamanoli-Siganidou, op. cit. (n. 3), 409, pl. 458 γ; Petsas, Makedonika cit. (n. 3), 297, pl. 11 α; JHS Archaeological Reports 1967–68, 16, Fig. 22; Brenk, op. cit. (n. 21), 248–250, pl. 83, 1, incorrectly identified as Hermes. That the figure is a Cabirus is clear from the drinking horn held in his right hand, and the traces of the attribute held in his left hand, above his left shoulder.

³⁶ The right-hand volute is restored, Karamanoli-Siganidou ArchDelt cit. (n. 3), 409, pl. 458 α; Petsas, Makedonika cit. (n. 34). 297, pl. 10 β; Brenk, op. cit. (n. 21), 284–250, pl. 80, 2.

³⁷ Karamanoli-Siganidou ArchDelt cit. (n. 3), 409, pl. 458 δ; Petsas, Makedonika cit. (n. 34) 297, pl. 11 β; Brenk, op. cit. (n. 21), 248–250, pl. 81, 1; BCH xcii (1968), 904, Fig. 7.

³⁸ Karamanoli-Siganidou, ArchDelt cit., 409:

³⁹ Makaronas, praktika, 1950, 312, Fig. 10.

⁴⁰ Seston, W., Diocletien et la Tétrarchie i (Paris, 1946), 217.Google Scholar

⁴¹ RIC vi, 710. See too, the panel on the Arch of Galerius where Zeus and Heracles are represented in relief on the altar in which Diocletian and Galerius are sacrificing: see H. P. Laubscher's forthcoming monograph (n. 22, above), under panel B II 17; Makaronas, Ή Καμάρα, Fig. 43; the clearest published photograph is Grabar, A., The Beginnings of Christian Art (1961), 148Google Scholar, Fig. 152. Zeus also appears in the background of Laubscher's panel B II 21; and Makaronas, op. cit. Fig. 34.

⁴² Gaebler, H., Die antiken Münzen von Makedonia und Paionia, 2. Abtg. (Berlin, 1935), 122Google Scholar, no. 25, pl. 23, 14.

⁴³ Edson, C. F., ‘Macedonica II, State Cults of Thessalonica’, Harv. St. Class. Philol. li (1940), 129–135.Google ScholarNow = IG x, 1, 2, no. 32. Cf. ibid. nos. 62, 259, 275 and 276.

⁴⁴ ‘Cults of Thessalonica (Macedonica III)’, Harv. Theol. Rev., xli (1948), 188–204.

⁴⁵ Edson, op. cit. 192–4. Now = IG x, 1,2, no. 199.

⁴⁶ e.g. Gaebler, op. cit. (n. 42), 127 no. 52, pl. 24, 12.

⁴⁷ De errore profanarum religionum II (ed. Ziegler, , Munich, 1953), 54.Google Scholar

⁴⁸ Heuzey, L. and Daumet, M., Mission archéologique de Macédoine (Paris, 1876), pl. 22 bis.Google Scholar

⁴⁹ Guerrini, L., ‘Las Incantadas di Salonicco’, Archeologia Classica xiii (1961), 55–8Google Scholar, pl. 21, 1.

⁵⁰ op. cit. (n. 43), 197–9.

⁵¹ Die Kabiren (1950), 210.

⁵² Laubscher's panel B II 21 (= Makaronas, Ή Καμάρα (cit. n. 24) Fig. 34). First recognized as Dioscuri by Reinach, S., Répertoire de reliefs grecs et romains i (1909), 390.Google Scholar

⁵³ Referred to by D. Kanatsoulis, Μακεδονική Προσωπογραφία, supplement (Thessaloniki, 1967), 31, no. 1767. It now appears that Μόρρυλος was a city in nearby Paionia; cf. E. Mastrokostas, Ἄγνωστοι πόλεις τῆς Παιονίας, Second International Symposium on Ancient Macedonia, Thessaloniki 1973 (forthcoming).

⁵⁴ The arch, complete: BCH lxxxii (1958), 757, Fig. 4; JHS Arch. Reports 1956–57, 13; Hoddinott, op. cit. pl. 9, b; Vlad Borelli, L., ‘Salonicco’ EAA vi (1965), 1084Google Scholar, Fig. 1196; Hanfmann, G. M. A., Classical Sculpture (1967), 339Google Scholar, Fig. 313; Grabar, A., Cahiers archéologiques xvii (1967), 78Google Scholar, Fig. 19; Papahadjis, N., Monuments of Thessalonike, 3rd edn. (Thessaloniki, 1968), 12Google Scholar; Bianchi Bandinelli, R., Rome, the Late Empire (1971), 305–6Google Scholar, Fig. 283. Galerius: Rüsch, A., ‘Das kaiserzeitliche Porträt in Makedonien’, Jdl lxxxiv (1969), 194–5Google Scholar, Fig. 109; Makaronas, op. cit. Fig. 48. Tyche: BCH lxxxii (1958), 757, Fig. 5; A.N. Letsas, Ἱστορία τῆς Θεσσαλονίκης (Thessaloniki, 1961), frontispiece; Makaronas, op. cit. pl. 47.

⁵⁵ IG x, 1, 2, no. 257.

⁵⁶ Karamanoli-Siganidou, , ArchDelt xx (1965), Chron., 409, pl. 457.Google Scholar They appear on the plans of the Octagon in Arch. Delt. xxii (1967), 388–9, Figs. 8–9, and in Makedonika ix (1969), pls. 1 and 2, between pp. 152 and 153.

⁵⁷ Mingazzini, P., L'Insula di Giasone Magno a Cirene (1966), 139–143Google Scholar, pls. 19, 4, 20–1. For the date of the house's destruction see Goodchild's, R. G. review of Mingazzini, Libya antiqua iv–v (1966–1967), 259Google Scholar, and idem. Kyrene und Apollonia (1971), 85, pl. 21.

⁵⁸ Aurigemma, S., I mosaici di Tripolitania (1960), 56 Figs. 136–9, 177Google Scholar, followed by Foucher, L., ‘Sur les mosaiques dc Zliten’, Libya antiqua i (1964), 9–20.Google Scholar

⁵⁹ Gnomon xxiv (1954), 111. This date is also preferred by Salomonson (see note 60).

⁶⁰ J. W. Salomonson, La mosaïque aux chevaux de I'antiquarium de Carthage (1965).

⁶¹ Becatti, G., ‘Mosaici e pavimenti marmorei’, Scavi di Ostia iv (1961), 230–1Google Scholar pl. 209, no. 429.

⁶² ibid., 27–9, pls. 209 and 219, no. 49. This panel and those from the House of the Nymphaeum are closest in style to those of the Octagon.

⁶³ ibid., 103–4, pl. 207, no. 189. The fine, but roughly laid, panels in the chancel of the sixth-century West Church at Apollonia in Cyrenaica are assumed by the excavators to have been robbed from an earlier structure; see W. M. Widrig and R. G. Goodchild, PBSR xxviii (1960), 78, pls. 27a and 32e. The same is presumably true in the case of the even more untidy opus sectile in the contemporary church at Ras el-Hilal; see Harrison, R. M., ‘A sixth-century church at Ras el-Hilal, Cyrenaica’, PBSR xxxii (1964), 5Google Scholar, 14, pl. 8b, c. In view of this, it would not be surprising if the opus sectile panels that formerly adorned the walls of the nave of the basilica of St. Demetrius in Thessaloniki (G. and M. Soteriou, Ἡ Βασιλικὴ τοῦ Ἁγίου Δημητρίου Θεσσαλονίκης. Athens 1952, pls. lb and 2a) were spolia from an earlier building. According to the scale drawings now preserved in the Courtauld Institute, they too measured 0.90 × 0.90 m.

⁶⁴ N. Papahadjis, op. cit (above, n. 54), 12.

⁶⁵ Makaronas, , Praktika 1950, 313–4Google Scholar (a very tentative suggestion); BCH lxxx (1956), 318; Hoddinott, op. cit. (n. 24), 124; Papahadjis, l.c. See now Kleinbauer, W. E., ‘The original name and function of Hagios Georgios at Thessaloniki’, Cahiers archéologiques xxii (1972), 56Google Scholar, who thinks that the Octagon may have served as the cathedral of Thessaloniki during the early Christian period, basing this notion on the ‘baptistery’ supposed to have been found outside it (on which see n. 73 below).

⁶⁶ See e.g. ‘plan restauré’ in Hébrard, E.—Zeiller, F., Spalato, le Palais de Dioclétien (1912), 70.Google Scholar

⁶⁷ Plan in Rivoira, G. T., Roman Architecture (1925), 218Google Scholar, Fig. 266 (after Bianconi, C.. Descrizione dei Circhi [Rome, 1789] pl. 1Google Scholar). A. Frazer has proposed that the Mausoleum was built by Maxentius for himself, even though his son was buried in it: ‘The iconography of the emperor Maxentius’ building in Via Appia', Art Bulletin xlviii (1966), 385–392.

⁶⁸ Ashby, T., ‘Classical Topography of the Roman Campagna’ PBSR i (1902), 156–161Google Scholar; Lugli, G., ‘La villa dei Gordiani e i monumenti al III miglio della Via Prenestina’ Bull. Com. xliii (1916), 160–163Google Scholar; Boethius, A. and Ward-Perkins, J. B., Etruscan and Roman Architecture (1970), 503–4Google Scholar, Fig. 191 and note p. 579.

⁶⁹ Deichmann, F.W. and Tschira, A., ‘Das Mausoleum der Kaiserin Helena und die Basilika der Heiligen Marcellinus und Petrus an der Via Labicana vor Rom’, Jdl lxxii (1957), 49Google Scholar, Fig. 5. cf. Deichmann, , Studien zur Architektur Konstantinopels im 5. und 6. Jahrhundert nach Christus (1956), 99.Google Scholar

⁷⁰ Stettler, M., ‘Zur Rekonstrucktion von S. Constanza’, RM lviii (1943). Beil. 1–3.Google Scholar

⁷¹ e.g. the plan in Hébrard, , BCH xliv (1920), 19Google Scholar, Fig. 8. See Grégoire, A., ‘La rotonde de S. Georges à Thessalonique est le mausolée de Galère’, Byzantion xiv (1939). 323–4.Google Scholar and Deichmann, F. W.. Ravenna, Geschichte und Monumente (Wiesbaden, 1969), 215.Google Scholar

⁷² He died and was buried at Serdica (Moreau, J., Lactance, De la mort de persecuteurs [1954] 396Google Scholar, for references). Perhaps it was considered that the body of Galerius (who had been ‘horrenda tabe consumptus’ Lactantius, De mort. pers. 35) would not travel well! It would, of course, have been impossible to move Galerius' body to Thessaloniki after Licinius' victory over Maximinus at the Campus Egenus in 313; Licinius, indeed, put Galerius' widow to death at Thessaloniki in 314; Gregoire, l.c. and Byzantion xiii (1938), 565.

⁷³ BCH lxxx (1956), 318. Hoddinott suggests on the strength of this ‘baptistery,’ that the Octagon may have been the church of the Asomati, attested in late Byzantine sources, but this is impossible.

⁷⁴ Makaronas, , Praktika 1950, 313–5Google Scholar, Figs. 11–13.

⁷⁵ As suggested by Swoboda, K. M., Römische und romanische Paläste (3rd edn. 1969), 303Google Scholar, n. 86.

⁷⁶ e.g. Wightman, op. cit. 106, plan.

⁷⁷ op. cit. 304.

⁷⁸ Vickers, M., ‘A note on the Byzantine palace at Thessaloniki’, BSA lxiv (1971), 369–71.Google Scholar

⁷⁹ Vickers, M., ‘The date of the walls of Thessalonica’, Istanbul Arkeoloji Müzeleri Yilligi xv–xvi (1969), 313–14Google Scholar; idem, ‘The Hippodrome at Thessaloniki’, BSA lxvi (1971), 369–371; idem, ‘The late Roman walls of Thessalonica’, Transactions of the 8th International Congress of Roman Frontier Studies, Cardiff/Birmingham, 1969 (forthcoming).

⁸⁰ The nature of the destruction fill—large fragments of rubble lying immediately above the marble pavement—perhaps points to an earthquake, and there are several potential candidates: (1) Ammianus Marcellinus xxiii, 1, refers to earthquakes in late 362 and early 363, but only mentions Nicomedia and Constaninople; Coronelli, V.. however, in his Epitome cosmografica (Cologne, 1693), 298Google Scholar, says: ‘nel 362 insorsero più robusti che mai in Costantinopoli e nello stesso tempo voltarono in Italia, passando per Tessalonica, che restò desolata’. (2) The ‘universal’ earthquake of 365, which caused destruction in places as far apart as North Africa, Palestine, Nicaea, Sicily, and ‘all the cities of Greece, except one’, Ammianus xxvi, 10; Libanius, Or. xviii, 291–3 (fully discussed by Goodchild, R. G., ‘A coinhoard from Balagrae, [El-Beida], and the earthquake of A.D. 365’, Libya antiqua iii–iv [1966–1967], 203–11Google Scholar). (3) and (4) Cedrenus recorded ‘universal’ earthquakes in 368 (ed. Bonn i [1838] 543–4), and at an uncertain date in the reign of Gratian, i.e. between 375 and 385 (ibid. 550–1). The latter is said to have destroyed a large part of Crete, Achaea, Boeotia, Epirus, and Sicily. (5) Marcellinus Comes, ed. Th. Mommsen (Berlin, 1894) 64, mentions an earthquake in the European provinces of Byzantium in November 394, and perhaps Thessaloniki was affected then. (6) Chron. Pasch, (ed Bonn 1832), 574, mentions an earthquake which was felt at Constantinople on Good Friday 417; could it be the same as the one which Coronelli, op. cit. 299, says ‘nel 416 fece strage in Tessalonica’?

I must express my thanks to Dr. N. N. Ambraseys, of the Department of Engineering Seismology at Imperial College, London, for many of these references.

Acknowledgements. In writing this article I have benefited from discussion with Dr. W. H. Plommer, Professor D. E. Strong, and Professor J.M.C. Toynbee, and from being able to read in manuscript Dr. H. P. Laubscher's forthcoming monograph on the Arch of Galerius. Any mistakes are my own. Dr. J. Christen kindly supplied Plate I, and Dr. B. Brenk Plates III, IV, V and VI, and Plate II was obtained from the German Archaeological Institute in Athens through the good offices of Dr. Laubscher. I must also acknowledge the assistance given to me in Thessaloniki by Dr. Ph. Petsas and Professor Ch. Bouras.