The Mint in Salona: Nepos and Ovida (474-481/2), in: "Studia numismatica Labacensia Alexandro Jeločnik oblata" (= Situla 26), Eds. P. Kos & Ž. Demo, Ljubljana 1988, 247-270. more

247 The Mint in Salona: Nepos and Ovida (474-481/2) Zeljko Demo The very last on the long list of emperors who governed the western part of the Roman Empire during the 5th century is Julius Nepos, a native of Dalmatia.1 Nepos' early career is entirely obscure but from when he finally appears in the literary sources on 1 June 473 as magister militum Dalmatiae (Cod. lust. VI, 61, 5), interested in problems of inheritance and family law, it took him a bit more than a year to reach the imperial purple. In such a short period Nepos completed his list of honours. After marriage to a relation of the Empress Verina, the wife of Leo I (457^74), a patrician title was added and when the situation in Constantinople was stabilized after the death of Leo I (January 18) he left for Italy in the early spring of 474. If the whole operation was backed by the eastern court, the necessary act of formal recognition would be openly manifested. This occurred at Ravenna where he first landed2 and was proclaimed Caesar under the procedure preceding the elevation to the rank of Augustus. Only on 24 June 474 did Nepos finally reach the top, having previously deposed Glycerius and expelled him to assume the position of bishop in Salona, capital of Dalmatia. According to the scarce historical evidence, the political activity of the new western ruler was directed toward Gaul where the Visigothic King Eurich (466-484) intended to complete a successful military campaign in the eastern parts of Aquitania I (Auvergne). When the Roman troops in Gaul could not succeed in the spring of 4753 Nepos was forced to conclude a peace. It may be assumed that the failure in southern Gaul and the change in Constantinople, where Basiliscus overthrew Zeno at the beginning of 475, weakened his position. When 1 Unless otherwise cited, for the historical background and dates referring to the period 473-481/2 see: J. B. Bury, History of the Late Roman Empire from the Death of Theodosius I to the Death of Justinian (A. D. 395 to 656), New York 1958 (Dover Edition), 343, 404-411, 417-421; L. Schmidt, Die Ostgermanen, Geschichte der deutschen Stamme bis zum Ausgang der Volkerwanderung, Miinchen 19692, 314-321, 334 f., 490 f. 2 On the basis of Jord. Rorn. 338 a thesis was elaborated by W. Ensslin s.v. 'Nepos' (6), RE XVI, 1935, 2506 but it was rejected by a number of historians mainly because of chronological obscurity related to that part of Jordanes' testimony where Domitianus' mission was attributed to the lifetime of the emperor Leo (I) and not to Zeno. See A. Demandt, s.v. 'Magister militum', RE Suppl. XII, 1970, 678 ff. 3 H. Wolfram. Geschichte der Goten von den Anfangen his Mitte des sechsten Jahrhunderts, Miinchen 19802, 226. 248 zeljko demo an unnamed enemy appeared in the summer of 476,4 the new patrician and magister militum Orestes was suddenly appointed to handle the case. The situation turned disastrous when Orestes moved against the emperor himself. Nepos left Ravenna on 28 August 475 and ascendens navem fugam petit ad Salonam (Anon. Val. 36, 5). From the period until another emperor fell and Odovacar seized Italy for himself, nothing is really known about Nepos, Salona or Dalmatia. Then, Nepos' name appeared again among those who were ready to follow the tradition of legislative rights rather than to see the ex-emperor spending the rest of his lifetime on the sidelines of historical events. In 476/477, an embassy sent to Constantinople gained formal recognition but not funding or military support for a full restoration. Only in 479 might hope have arisen when Theodoric offered to lead his Ostrogoths into Dalmatia and fight for Nepos' cause. One year later nothing was as interesting for ancient chronicles as Nepos' violent death haud longe a Salonis, sua in villa (Marcel, com., s. a. 480) presumably Dioletian's ancient palace.5 Actually it was a treacherous murder said to have been instigated by Glycerius but performed by Nepos' comites Ovida and Viator, the former of gothic or eastern german origin.6 This plot, carried out on 9 May 480, was not answered by any articulate response for quite a time. Only in late 481/4827 did Odovacar in Dalmatiis Ovidam vincit et perimit (Cass. Chron., s. a. 481). From the numismatic point of view certain historical events and policies during the various periods of Nepos' reign in Italy and Dalmatia might indicate the adoption of a mint practice in the emperor's native province. Regardless of when it happened, this was an act of imperial authority with substantial administra- tive, fiscal and monetary consequences for the entire provincial economy. The idea is of great importance for the history of Dalmatia but the possibility that Nepos' coinage includes the products of a mint located in Salona was not made explicit until very recently.8 G. Lacam made the first attempt in assigning several 4 For the most appropriate explanation of these 'hostes' (Jord. Get. 241), see J. M. O'Flynn, Generalissimos of the Western Roman Empire, Edmonton 1983, 195 n. 26. 5 The thesis is not later than the mid 18th century (D. Farlati, Illyricum sacrum, T. II, Venetiis 1753, 125.2) but has been accepted on many occasions subsequently (E. Hebrard - J. Zeller, F. Bulic-J. Bervaldi, G. Novak, E. Dyggve, F. SiSic, B. Saria, J. Wilkes, etc.). The most accessible arguments could be provided by the ancient sources of the late fourth or fifth century (Eutropius, Jerome, Prosper Tiro) where Diocletian's palace, with or without the name Spalato or Aspalatho, was always specified as villa. . haud longe a Salonis. For the name of Diocletian's palace and related problems, see F. Bulic - Lj. Karaman, Palaca cara Dioklecijana u Splitu, Zagreb 1927, 11-15. 6 E. Forstemann, Altdeutsches namenbuch I Personennamen, Bonn 1900, 1195; M. Schonfeld, Worterbuch der Altgermanischen Personen-und Volkernamen, Heidelberg 1911, 179. 7 Plre II, 815. 8 G. Lacam, La Fin de Vempire romain et la monnayage or en Italie, 455^193, Luzern 1983, 596f., 604 f. and 607 with PI. CXLVIIa (Type 1, Variete a/b; Type 2) or PI. 37:1-3 [Salone (?)], 629 f. and 631 with PI. CLIV (Groupe 1 Type 1) or PI. 38:21 [Salone pour Milan (?)], 703-713 including Pis. CLXXVII-CLXXIX or PI. 40:58-62 [Emissions de Nepos en exil. Salone], and PI. 46:168-173 [Salone. Emissions de Nepos en exile au nom de Zenon (solidi and tremisses)]. Corresponding to. Lacam's general distinction of three minting periods of Nepos, the Salonitan coinage is divided into three groups (p. 596 f.). The most interesting is the first one represented by three solidi said to have been minted before he left for Italy in 474 (p. 604 f.). The other two are of later date, rarely attributed to Nepos' actual reign 474-475 (p. 629 ff.) but predominantly to his nominal second reign 476-480 (p. _the mint in salona: nepos and ovida (474-481/2)_249 fairly different groups of Nepos' coins (almost exclusively solidi) to the mint in Salona. He did not discuss tremisses but three examples in the name of Zeno were also attributed to Nepos and Salona(?). However, the present chronology of these issues precluded acceptance of the fact that a certain number of coins (solidi and also tremisses) were the product of a continually productive Dalmatian mint (474-480). In the absence of such a hypothesis it was impossible to establish a consistent pattern according to which groups or issues could be collated and organized into workable order. Starting from this premise, the final conclusion may be quite different. Moreover, the numismatic evidence although still incom- plete suggests that minting continued even after Nepos' death during the relatively short reign of Ovida (480-481/2). NEPOS (474-480) Tremisses All the coins bear a western cross-in-wreath reverse with COMOB in the exergue. The weight when given ranges from 1.40-1.50 gr. and the axis | when known or detected (*) is the rule without exception. The lettering t, for L and D for D appear only in individual groups while E for E represents a truly significant and very remarkable element of the entire issue. At least four obverse inscriptions may be noted: 1) DNIVUSfE POSPE/VC, 2) pNIVWE POS- PFAC, 3) DMIVLNE POSPFAVC, 4) DNIVLNE POSPFAVC. All thirteen examples known to me were minted in five obverse and five reverse dies. These form five quite different groups. Group I can be denoted as »Milan« on the basis of style and legend. However, differentiating anomalous elements is not difficult. The wreath on Milan tremisses drawn as fourteen pairs of small spikes (PI. 1:3) was replaced here by a combination of 14/15 (left/right). Even on poorly preserved specimens the portrait with its oversized nose can easily be detected. The lettering l, for L is notable but E for E in the first half and E for F (!) in the second half of the obverse inscription are significant parts of the legend DNIVLNE POSPE/VC which have attracted attention since the days of A. Banduri9 or H. Cohen.10 PE/VC seems to be nothing other than a transformed ending of the very early and rare Milan PE/VC (PI. 1:2). The lettering E for E represents a distinctive marker of this coinage present in every group and every type. It also appears on Glycerius' 702-713). Neither of these two groups can actually be considered to be of Dalmatian origin. - For the main criteria distinguishing the two minting periods of Nepos, see the instructive article of J. P. C. Kent, 'Julius Nepos and the Fall of the Western Empire', Romische Forschungen in Niederdsterreich V (= Corolla Memorie Erich Swoboda dedicata), Graz-Koln 1966, 149 f. 9 Numismata imperatorum romanorum a Traiano Decio ad Paleologus Augustus, T. II, Lutetiae Parisiorum 1768, 603. 10 Description historique des monnaies frappees sous I'empire romain, V" ed., VI, Paris 1862, 531 No. 11= 2™ ed., VII, Paris-London 1892, 241 No. 18. 250 zeljko demo tremisses (PI. 1:6) as well as Nepos' tremisses and V4 siliquae minted in Ravenna in 474-475 (PI. 1:9,10). After O. U. Bansa included one coin of this group among tremisses attributed to the mint in Milan (No. 2),11 G. Lacam recently published another two, assigning them to his Milan Type 4/Variete A dated in the period of Nepos' actual reign (Nos. 4-5).12 To the published examples from the museum collections in Paris and La Haye, a tremissis from NMLj (No. 1) - previously from the Pokrajinski Muzej in Maribor - has been added (Fig. 1). NMLj Bansa 1949, Nr. 169 = J. Schulman 31/05/1938, Nr. 528 = MuM 38/1968, Nr. 499 = Lacam, 689, 687, PI. CLXXIIL2-3. Lacam, 689, 687, PI. CLXXIIL1 (BN) Lacam, 689, 687, PI. CLXXIIL4 (La Haye) [x Lacam, 689, 687, PI. CLXXIIL6 (coll. prive) 1.42/13.5/ | 1.40/ - / 4 1.42/14 I - 1.44/13.8/ | * 1.43/13.5/ - ] GROUP II introduces an obverse wholly new and not at all similar to the »Milan« style of group I. Considering the decoration on the top of the diadem, the model was taken from Nepos' Roman tremisses (PI. 2 :16). In addition to the previously known U for L and particularly E for E, the lettering D for D was added to the obverse legend DNIVWE POSPFAC. The abbreviation AC (for AVC?) has a good analogy in the several years earlier tremisses of Anicius Olybrius (PI. 1:5) assigned to Rome,13 but the ending PFAC, perhaps known to H. Cohen,14 is attested on a tremissis of Nepos minted in Arelate.15 11 O.U. Bansa, Moneta Mediolanensis, Venezia 1949, 326, 327 No. 169 (Tav. XIV: 169). 12 G. Lacam, o. c. (n. 8), 689 Nos. 1-6, with PI. CLXXIII: 1-5 or PI. 41:85-87. His example No. 6 (see here No. x) is not documented with a picture, while No. 5 must in any case be excluded. Note that Lacam's Nos. 2 and 3 are the same coin (see here No. 2). 13 O. U. Bansa, o. c. (n. 11), 288 and corresponding n. 74, Tav. N:b. 14 O. c. (n. 10), \" ed., 531 No. 10 = 2™ ed., 241 No. 17. 15 G. Lacam, o. c. (n. 8), 681 with PI. CLXX: Type 1 (Roma). _the mint in salona: nepos and ovida (474-481/2)_251 The obverses of all three examples of Group II were minted in the same die. This is recognizable from the vertical hollow reproduced on the coins as an extrusive perpendicular line extending from the upper part of the bust of the emperor across the forehead and nose to the letter P at the beginning of the second part of the obverse legend. The reverses were minted in two dies. Type A with the »Milan« reverse of Group I is represented by two examples both presently in the collection of AMS. One of them (No. 6) was found in 1920 near Vodice, ca. 60 km west of Split (Fig. 2). 5 AMS 125 1.46/ 13 / j 6 AMS 344 (near Vodice, 1920) 1.44/ 13 / j No. 5 No. 6 Fig. 2 Type B has a reverse equivalent in style to tremisses minted in Ravenna under Odovacar but in the name of Zeno (PI. 1:14). Note the small dot under the cross, two dots above the letter B in the exergue and particularly the number 252 Zeljko demo of spikes in the wreath (14/15) identical to that on the »Milan« designed reverse of groups I and IIA. The single example, »provenant d'une veille et celebre collections appeared at an auction in Paris just before the first World War (Fig. 3)."- 7 Bourgey 16/12/1913, Nr. 774 - / - / - Fig. 3 No. 7 Group III shows the final step in the creation of a constantly new obverse style. The portrait bust and the top of the diadem bear a resemblance to the tremisses and V4 siliquae minted in Ravenna during Nepos' actual reign (PI. 1:9,10). In the obverse inscription DUIVLNE POSPFAVC the lettering k for L, ligature /VC or the abbreviation AC have been removed. An unexpected novelty is the form N for N, while E for E remains unchanged. The single coin (Fig. 4) unfortunately exhibits overstriking of its obverse and apparent traces of a fairly long used reverse die of the groups IA and IIA. 8 AMS 189 1.44/ 13 / 4 Fig. 4 On account of the reverse type, confirmation is required. _the mint in salona: nepos and ovida (474-481/2)_253 Group IV presents the same obverse die as the previous group III. The only difference can be found in the accurate and confident obverse inscription DNIVLNE POSPFAVC where the initial M of group III was recut into its regular form. Such an unusual step marks the point where a progression in the style and concept of the obverse die finally lost all meaning. The four coins that make up Group IV were all of the same obverse die. This one was used together with three different types of reverse dies. Type A is the same long used die cited in groups I, IIA and III (Fig. 5). The only example of this type, attributed to Milan, has the heaviest weight standard of the entire issue. The coin was published over ten years ago as a find from Pazin (Pisino) in Istria.17 9 Gorini 1974, No. 12 (Pazin ?, Istria) 1.50/ 13 / J, Fig. 5 type B is the first in a series of similar and fairly numerous reverse types (Groups IVC and V) that copy the concept related to the tremisses minted in Ravenna during Nepos' actual reign (PI. 1:9). It may also be of a certain importance that from this point on both obverse and reverse dies will correspond to the style of the same Italian mint (Ravenna) as had already happened in the minting period of group I (Milan). The single example of this type depicts the wreath as fourteen pairs of parallel spikes (Fig. 6). 10 Reichmann XX/1922, No. 1481 1.47/14.7/ - 17 G. Gorini, 'La collezione di monete d'oro della Societa istriana di archeologia e storia patria', AttMemSIASP XXII, 1974, 115 No. 12, 178 f. 254 zeljko demo No. 10 Fig. 6 Type C merely illustrates a variation of the previous reverse type. Along with a small imprecision in the left section, the wreath was drawn as twelve pairs of parallel spikes. Of two specimens (Fig. 7), the one from the former collection of Fr. Imhoof-Blumer was sold in Miinchen at the beginning of this century (No. 11) and the second, cautiously attributed to the mint in Rome, appeared in Zurich more than ten years ago (No. 12). 11 Hirsch 18/1907, No. 1770 12 Sternberg 1975, No. 524 No. 11 No. 12 the mint in salona: nepos and ovida (474^81/2) 255 Group V is represented by only one coin which is simultaneously the single example not die-linked to any of the previous groups and types. The portrait, generally poor and carried out in a somewhat shallow relief, is directly dependent upon the style of Groups III and IV but the lettering E for E is left unchanged. The reverse, although similar to Groups IVB and IVC, has a wreath drawn as ten pairs of parallel spikes. The single example, once in the collection of the Franciscan monastery in Fojnice (ca. 50km west of Sarajevo), has been in the ZMS collection since 1901. More detailed data about the provenience are not cited and the coin is noted only as a find from Bosnia (Fig. 8). 13 ZMS 832 (Bosnia, before 1901) 1.45/ 13 / j Fig. 8 All thirteen tremisses divided here into five groups and five types have the usual cross-in-wreath reverse introduced in the West soon after the arrival of Galla Placidia (May 425) and the proclamation of Valentinian III (23 October 425).18 From then on this was retained as a recognizable sign of Italian mints to the last decades of the 5th century, thus not only after the first promising successes of Theodoric's Ostrogoths as an early sign of the new politics of the future ruler of Italy. In Gaul, mostly controlled by various German tribes, the cross-in-wreath was very popular during the reign of Valentinian III but it began to be abandoned at the end of Majorian's reign and particularly under Libius Severus. In the very few cases where this did not occur, as is shown by rare examples of Anthemius and Nepos, the responsibility was taken by a local ruler or some other monetary authority still influenced by a western emperor or political affairs taking place in Italy. This probably gave inducement to Nepos in late 474 to renew minting of solidi with the mintmark AR in the field left or right of Victory (PI. 2:18). These are followed by rare tremisses with a few variations of an always unbroken obverse legend (DNIVLNEPOSPFAC, DNIVLNEPOSPFAVC)19 and the cross-in-wreath 18 The first tremisses of the cross-in-wreath type in the name of Valentinian III were said to have appeared in 426: J. P. C. Kent, 'Un monnayage irregulier du debut du Ve siecle', CENB 11, 1974, 27. 19 G. Lacam, o. c. (n. 8), 680, and 681 with PI. CLXX: Type 1 (Roma); A. S. Robertson, Roman Imperial Coins in the Hunter Coin Cabinet, V. Diocletian (Reform) to Zeno, Oxford 1982, 464 No. 4, PI. 91: 4 (uncertain mint). 256 zeljko demo reverse similar in style to those of Majorian or Libius Severus. Tremisses minted in Arelate as well as those created in all three main Italian mints bear the contemporary style of their minting locations. The thirteen examples described above, despite various combinations and influences, show a quite different regional style. The sites or collections where the majority of this coinage was either found or preserved lead to the conclusion that the eastern provinces of praefectura Italiae were their minting place and circulation area. In the second half of the fifth century, Dalmatia and Salona as its capital played a considerable role in western political affairs. Nepos was a native of this province, here he began his Italian campaign in the early spring of 474, here he sent Glycerius to assume the vacant position of Salonitan bishop, and it was here he finally returned after failure in Italy and his departure from Ravenna on 28 August 475. On the basis of contemporary mint practice it may be assumed that coins would be issued in the city of imperial residence.20 This is exactly what Salona was to become a year later when the news of the murder of Orestes and the fall of Romulus Augustulus reached the Dalmatian coast in September 476. The attribution of all Salonitan tremisses to the period 476-480 encourages the connection of this coinage with events affecting just a part of Nepos' political life. This hypothesis seems to be contradictory to the political activity of Nepos during his actual reign in 474-475. There is no factual or substantial evidence whatsoever to enable one to establish an earlier initial date of minting for these tremisses. The entire scale of stylistic evidence even indicates a slightly longer developmental process whose origin and route led from the confused and complicated opening issues to the clarified and standardized one of the final stage. This undoubtedly shows the existence of at least two classes of Salonitan coinage while the large number of groups and types as well as variations in their stylistic concept echo the different political aspects of the period. In one way or another almost every described group is perfectly die linked. This suggests that minting was maintained without interruption from the initial date at the beginning of Nepos' reign up to his violent death in May 480. Any striking in the name of Nepos before his arrival in Italy in the spring of 474 must be excluded. If minting started in 474 at the beginning of Nepos' actual reign (sometime after June 24 or even later during the same year), the present state of groups and types would certainly be incomplete. In spite of this, there is a distinct chronolo- gical order based on the variable influences summarized in the stylistic concept of each group. The first one (Group I) strictly followed an early Milan emission but was meant to be distinguished from its prototype in terms of a very small number of details (modification of the end of the obverse inscription, number of spikes in the wreath). The implications of the last seem to be directly related to the nature of the newly opened Dalmatian monetary office which started to work as a subsidiary mint basically dependent upon the style and type of Milan. Such 20 The placement of Nepos' mint in the capital of Dalmatia is suggested by the presence of the praepositus thesaurorum in Salona, which was mentioned in Not. Dig. Occ. X. This does not exclude the possibility that the mint could somehow have been developed within the Emperor's actual residence (supra n. 5). the mint in salona: nepos and ovida (474-481/2) 257 a position had to conform to the limitations imposed by contemporary administra- tive practice and for this reason could not endure. The different obverse style of the following few groups (II, III, IVA) combined with two differently influenced reverses gave a quite new character to the tremisses minted in Salona. Fairly distinctive from any other official issues, Italian or Gallic respectively, they experienced a developmental progression which significantly extended beyond the provincial boundaries and necessarily led to a number of institutional changes in the administrative status of the Salonitan mint. This may be a tangible part of a process started in late 474 or early 475 consisting of the restoration of the administrative, fiscal and monetary units within the remains of what had once been the dioecesis Illyricum in partibus occidentis. The attempt, more theoretical than real, certainly lost ground in 475 (after August 28) but the Salonitan coinage was not affected before 476/477. A significant change in the choice of types for both obverse and reverse appeared subsequently when tremisses of Groups IVB, IVC and V were cut in the style of Ravenna. This seemingly coincides with the results of some kind of agreement set up between Zeno-Odovacar/Roman Senate and Zeno-Nepos sometime after the successful mutiny of Odovacar's troops against Orestes and the fall of Romulus in the late sumer of 476 (September 4). If Zeno on the one side in his answer to both Odovacar's and Nepos'embassies continued to pretend that Nepos was still western emperor, Odovacar on the other was explicit in limiting his own role by mentioning patriciate and restricting his effective influence to within the boundaries of ttqv ^wv TtocAojv . . . StotxTj<jK2 (Malch. frag.10 = IV, 119).2] Dalmatia was evidently left aside as an independent unit under exclusive imperial control. The decision would be exceptional if it were not followed by an appropriate legislative explanation. This could be the moment when the province by being formally given or officially relinquished to a member of the imperial family achieved the status of patrimonium.22 This gave Nepos enough space to implement most of the imperial prerogatives. One of these certainly comprised the »transfer« of the Ravenna style to the tremisses minted in Salona. From this point of view the sequence and dating of Salonitan tremisses may be classified into the following periods: 1. Tremisses of »Milan« style (Group I) with /V in the ligature were struck at least contemporaneously to the Milan tremisses and solidi bearing the same ligature at the end of the obverse inscription (PI. 1:2,1). They all belong to the very early period of Nepos' actual reign but if Milan tremisses and solidi showing the same element are really so scarce the ligature N should have been abandoned in Milan before the end of 474. The subsidiary character of the Salonitan mint suggests the same date, between June/July 474 and late 474 (p. 250, Fig. 1). 21 For an interpretation of »diocesi degli Itali« in the sense of praefectura Italiae see: C. Cipolla, 'Considerazioni sul concetto di stato nella monarchia di Odoacre', Rendiconti della Reale Accademia dei Lincei, classe di scienze morali, storiche e filologiche XX5 1911, 384. 22 On the patrimonium in Dalmatia during Odovacar and Theodoric see: L. Schmidt, o. c. (n. 1), 324; A. H. M. Jones. The Later Roman Empire 284-602, A Social Economic and Administrative Survey, Oxford 19732 , 25 5 f. 17 Situla 26 258 z.eljko demo 2. Tremisses with an obverse free of Milan influence (Groups II-IVA) directly follow the previous period but the new stylistic concept of the various obverse types implies a conscious policy to separate the Salonitan tremisses from the actual products of the other three major mints in the Italian prefecturae. The number of groups and types is slightly increased, probably due to the longer period of minting these tremisses. They may cover the period between late 474 or early 475 and the beginning of the lustrum on September 1, 477 (p. 251-3, Fig. 2-5). 3. Tremisses of »Ravenna« style (Groups IVB-V) were introduced as a part of the same political compromise that gave inducement to the nominal »second« reign of Nepos. The choice of style indicates a product determined by government directed monetary policy, while the maintenance of the central mint authority contributed to the unified types of the obverse and reverse dies. The number of groups indicates another longer term of minting, presumably one that started on 1 September 477 and ended with Nepos' death on 9 May 480 (p. 254-5, Fig. 6-8). SOLIDI It was the usual practice in all major western mints in the second half of the 5th century that the production of solidi was accompanied by corresponding tremisses. Consequently, if Nepos minted tremisses in Salona it would be logical to conclude that solidi were also minted in the same facility. It could immediately be expected that finds from Dalmatia or adjacent regions would support such a thesis in an adequate manner. In this case it did not occur for a very simple reason: in the museum collections which offered such useful data about the Salonitan tremisses not even a single solidus of Nepos may be found. Lacking such a context it would hardly be possible to attribute to the mint in Salona any of the very few »unusual« solidi of Nepos which have appeared side-by-side at the same auctions with his Salonitan tremisses (PI. 2:20,21). All solidi minted for Nepos in either of his two minting periods bear the same reverse type: Victory standing left, holding a long jewelled cross, while COMOB and VICTORIA AVGGG are followed by the mint-marks RV, MD or AR for Ravenna, Mediolanum or Arelate. Curiousity is aroused by the problem of the »moneta palatii« attributed to Rome,23 where the mint-mark RM has been replaced by a star in the field right of Victory. In addition to evidently regular products of these four major western mints of Nepos, there is a certain number of fairly distinctive solidi usually bearing MD, or rarely RV, but minted elsewhere than Milan or Ravenna. Even more rare than these are specimens with a star in the right field. The considerable number of Salonitan tremisses have furnished many useful elements which can aid in recognizing the products of the Salonitan mint among the not scarce »similar«, »irregular« or »unusual« solidi minted for Nepos elsewhere than in Italy. The mixture of various influences so often seen on Salonitan tremisses provides the main criteria but the general style and lettering, O. U. Bansa, o. c. (n. 11), 316. _the mint in salona: nepos and ovida (474-481/2)_259 especially E (for E) and I, (for L) may be of great interest. A basic distinction between solidi as international currency and tremisses as a valuable but fairly local denomination cannot be avoided. With very few exceptions (Group IIB), tremisses of the Salonitan mint are distinguished by two basic groups of reverses: one designed in a style contemporary to Milan which may be related to Nepos' actual reign (Groups I, IIA, III and IVA), and a second with a reverse in the style of Ravenna adapted for the Salonitan mint after Ravenna had already abandoned this type during the reign of Romulus (Groups IVB, IVC and V). On the basis of the reverse type it could be expected that solidi would be influenced and designed according to the same propositions: 1) The »earlier group« should follow the basic elements of the Milan style which includes both MD in the field left and right of Victory and a star at the end of the reverse inscription as well as the position of the axis, f for solidi as it was | for tremisses. If the dies were cut in Salona, the solidi should contain all the elements mentioned above but should be distinguished by the style and the type of lettering. Seemingly, nothing of this kind has yet been published, or at least is not known to me. 2) The »later group« should be dependent upon the style and type of Ravenna. In 477, when the group is supposed to have come into use the choice of types was increased. However, it may range from the type with the mint-mark RV in the field left and right of Victory common to the period of Nepos' actual reign (PI. 1:7,8) to the type with an eight-pointed star in the field to the right of Victory introduced and minted in Ravenna for Basiliscus and Romulus after 28 August 475 (PI. 1:11,12). The contemporary monetary practice as well as the bulk of various political circumstances would strongly mitigate against both solutions if the star type were not simultaneously reminiscent of the »moneta palatik where a star replaced RM at the very beginning of Nepos' actual reign. If at that time the Roman mint could adopt the new reverse type as the simple sign of »moneta palatii«, then Salona might be playing the same role ca. 477 as well as in the following three or four years up to the death of Nepos in May 480. Any solidus of Nepos with a star in the field right of Victory, either of the early Roman Nepos type or the later Ravenna Romulus type, classified in terms of style as »similar«, »unusual« or »irregular«, could potentially be of Salonitan origin. The number of coins available for examination is actually very small. It consists of no more than three solidi, all of them already discussed, attributed and dated as well. The first one has been known for a long time (PI. 2:20) attributed either to the mint in Rome and dated to 474-47524 or lately to Salona but placed dans le hate de preparatifs pour le grand depart,1^ ie. even before Nepos actually seized the Western throne in the early spring-summer of 474. Although 24 O. c, 317, Tav. 0:g. This coin was also in the following auctions: J. Hirsch XXIV/1909, No. 2581 (coll. Weber) = J. Hirsch XXXI/1912, No. 2028 = J. Schulman 5/3/1923, No. 2925 (coll. Vierordt) = Leu - Numismatic Fine Arts 16-18/10/1984, No. 384 (coll. Garrett) = Numismatic Fine Arts XX/1988, No. 606. 25 G. Lacam, o. c. (n. 8), 605 and 607 with PI. CXLVIIa (Type 2) or. PI. 37:3 [Salone (?)]. 17' 260 zeljko demo this reverse type was introduced in Rome at the beginning of Nepos' reign (474) the coin is certainly neither from Rome nor as early as it is said to be. The general style is rather that of Ravenna solidi minted for Odovacar in the name of Zeno in the second half of 480 and sometime after (PI. 1:14). Some doubts may arise concerning the style of lettering if it is compared with that of the later Nepos tremisses of Salonitan Groups IV and V while the omission of the typical Salonitan lettering E for E cannot be ignored. However, as long as it is the only specimen of this kind as well as the only known pair of dies the attribution will alternate between Rome (hardly) to Ravenna (the most probable). The possible date will certainly not be earlier than that of the nominal »second« reign of Nepos (after late 476). The »Salonitan thesis« could come into question partly by being dependent upon the hypothetical but non-existent »earlier group« of Salonitan solidi and partly because of the quite different direction of developmental progression of the following two specimens. These are the best possible examples of the »similar«, »unusual« or »irregular« classes of solidi. One of them, said to be preserved in the museum collection in Milan (Fig. 10), has been known for a long time and was described as a solidus of the Visigothic King Eurich (466-484).26 The other one (Fig. 9) attributed to the mint in Rome, was probably that mentioned as a provenienced find from Oland in Sweden (Koping sn., Kullen).27 Recently, both were correctly assigned to the mint in Salona.28 The presence of E for E in the obverse inscription may only confirm the attribution while the general style of lettering leaves no doubts that the same hand which engraved the obverse die for Salonitan tremisses of groups IIA and IIB (p. 251, Fig. 2) was involved in preparing the dies for the solidi. Happily both coins were not exactly the same. A few minor distinctions in the formation of,the reverse types are of great interest. They throw an entirely different light on the structure and practice of minting Salonitan solidi, suggesting that very little of the previous division into »earlier« and »later« groups deduced from the developmental progression of the reverse types of Salonitan tremisses would be applicable to solidi. Both Salonitan solidi of Nepos bear the same obverse inscription DNIV WE POSPFAVC but the variously shaped star in the field right of Victory and the end of the reverse inscription with or without pellets provides more valuable information. The last criteria is seemingly more important as it defines two chronologically diverse groups: 1. VICTORI AAVCCC, star in the field right of Victory, (Fig. 9). 26 W. Reinhart, 'Die Munzen des tolosanischen Reiches der Westgoten', DJN 1, 1938, 120, 132 No. 104, Taf. 6:104.' H/0V~*w. t?X0, i ■ *V» vi* ^ , Mb , - 1\ « ■■■' "■• •• * «») 27 J. Fagerlie, Late Roman and Byzantine Solidi Found in Sweden and Denmark (= NNM 157), 1967, 28 No. 180 (p. 188 No. 79 with older literature). 28 G. Lacam, o. c. (n. 8), 604 f. and 607 with PI. CXLVIIa (Type 1. Variete a/b) or PI. 37:1-2 [Salone (?)]. the mint in salona: nepos and ov1da (474-481/2) 261 Fig. 9 The omission of pellets was not uncommon in Italy in 474-476. They are lacking only on the earliest solidi of Nepos minted in Ravenna (PI. 1:7) and subsequently on a few occasions during the reign of Romulus, on a solidus minted in Arelate (PI. 2:19), on a few specimens from Milan (PI. 1:4)29 as well as on those that were considered to be of Roman origin (PI. 2:17). All Nepos' solidi from Arelate minted between 474-475 (and after 476)30 are also of the type without pellets (PI. 2:18). If the pellets in Ravenna or Rome and the star in Milan could be explained as the distinguishing mark of Nepos' Italian mints, the omission of any kind of sign at the end of the obverse inscription, pellets or star respectively, would indicate products of Nepos' regular mints working outside Italy, in Gaul (Arelate) and Illyricum [Dalmatia, (Salona)]. The omission at the end of the reverse inscription also serves to distinguish this group of solidi from the contemporary specimens minted in Rome during Nepos' actual reign (PI. 2:15). This suggests an early date for all Salonitan solidi without pellets. They correspond to the tremisses of groups IIA and IIB where the Roman obverse was also introduced, but on the basis of the lustrum cycle solidi of this type would be kept unchanged for some time afterwards. They date from late 474 or early 475 to (?) 1 September 477. 2. VICTORI AAVCCC:, star in the field r. of Victory, (Fig. 10) The pellets at the end of the reverse inscription were introduced at thex beginning of the new lustrum in September 477 while the choice of type is 29 The coin from AMZ (Inv. no. 25474) is die-linked to the specimen in the collection of Dumbarton Oaks (Washington) = R. Ratto 14/11/1907, No. 1514 = J. Hirsch XXXIII/1913, No. 1503 = Naville 11/1922, No. 1861. [Also in: M. Bernhart, Handbuch zur Munzkunde der romischen Kaiserzeit, Halle 1926, 10 No. 24 (Taf. 23:24); G. Lacam, o. c. (n. 8), 732 and 731 with PI. CLXXXIII (Type 1/Type 2. Variete a); M. F. Hendy, Studies in the Byzantine Monetary Economy c. 300-1450, 718 No. 12 (PI. 11:12)]. 30 The possibility was noted in: J. P. C. Kent, Roman Coins, London 1978, 334 No. 766. 262 zeljko demo Fig. 10 predominantly related to the formal establishment of Nepos' new palatium in Dalmatia after the actual recognition of his nominal »second« reign. The type was contemporary to the tremisses of groups IVB, IVC, and V and was undoubtedly current at the time of Nepos' death on 9 May 480. OVIDA (480-481/2) Tremisses Only five coins minted in the name of Zeno represent a set of tremisses which can be easily detected by their poor cut and generally simplified portrait style. The decoration on the top of the diadem (one or two pairs of dots) may attract some attention but the bust formed of little triangles is a remarkable element perhaps derived from regional attire. The lettering E for e, previously seen on the Salonitan tremisses of Nepos, is of great importance. It can be found in the obverse inscription which takes two forms (with and without a break): 1) DNZENOPERPT All (for N C) and 2) DNZ3NO PERPTNC. The cross type reverse with comob in the exergue shows that tremisses of this kind, in spite of details, tended to be included in the western monetary circle. However, they are all of regular weight, mostly 1.48 with 1.55 gr. noted in only one case. When it is given or can be reliably established, the axis f seems to be the rule. Two obverse and three reverse dies can be classified into the following groups: Group I includes three coins of the same obverse die with Zeno's name in an unbroken legend. The lettering E and E for e in the obverse inscription DNZENOPERPT AI ( suggests the minting place as well as the chronological po- sition of the entire group I. PERPT is certainly local but is too exceptional to _the mint in salona: nepos and ovida (474-481/2)_263 be easily explained as an extended form of PERP for perpetuus used simultaneous- ly on eastern solidi and tremisses. Al( is the only ill-formed part of this obverse which die linked two reverse types. type A has a wreath drawn of 13/11 (left/right) small spikes. Its binding is not particularly clear, but is undoubtedly very similar to the tie on the monogram type minimi of Theodosius II (PI. 2:25). Note the border made of beads moved closer to the wreath and attached to both ends of COMOB. The single example of this type was purchased at some point in Paris31 by Count Tolstoi for his once famous collection (Fig. 11). 1. Tolstoi, No. 56 1.55/ - I- Fig. 11 type B has a wreath depicted as twelve pairs of small spikes. The exergue line with a wreath around the cross looks similar to the representation so often seen on the eastern monogram type minimi in the second half of the 5th century (PI. 2:26). In addition, the two thick ribbons at the juncture of the exergue and the wreath connect this with the border of beading. Two coins of this type can be listed here but both are without provenience (Fig. 12): one was bought in 1893 for ZMS (No. 2) and the other is in the collection of AMS (No. 3).32 2. ZMS 1845 (purchased from J. Perer, 1893) 1.48/14-13/ f 3. AMS 142 1.48/ 14 / \ 31 This is the probable reason that the coin was described as a coarse copy minted in Gaul. The obverse inscription was misread as DNZCNODCRPTAVC. J. Tolstoi, Monnaies Byzantines I, S. Petersburg 1912, 155 No. 56. 32 A previous inventory number of the Archaeological Museum in Split (A 661) is still visible on the coin, but it was not possible for me to decode its meaning. Group II exhibits no difference in general style from the preceding specimens of group IB. The lettering 3 for E in the first half and the already seen E for E in the second half of the broken obverse inscription DNZ3NO PERPTA/C are most remarkable. The reverse was kept almost unchanged except for the number of small spikes in the wreath which were increased from 12 to 16 pairs. Only two coins can be noted (Fig. 13). One was purchased for the AMZ sometime in the last decades of the 19th century (No. 4) and the other (No. 5) appeared at an auction in Luzern in 1970 listed under »Ostgoten in Italien« and attributed to Odovacar (476-493). 4. AMZ 284 1.43/13.5/ f 5. Hess45/1970, 730 1.48/13.5/ f Coins of both groups are very rare and are exceptional in their choice of stylistic elements. They show a mixture of various influences, both western and eastern. The unbroken legend of group I is certainly more common for western tremisses (and solidi) produced in the second half of the fifth century than the mint in salona: nepos and ovida (474-481/2) 265 contemporary issues minted in the east. The long thin line of the exergue is rather eastern as is PERPT in the obverse inscription used on the unusual tremissis where Leo I (or II) is depicted (PI. 2:23). The cross type reverse with COMOB in the exergue is reliable evidence for a western issue but every other detail is hardly similar to any coinage minted in Zeno's name in Odovacar's Italy or elsewhere in the West (Gaul). All of this could represent an obscure and confused document of a fairly autonomous monetary authority practising and minting somewhere between the East and the West. Tremisses in the museum collections in Sarajevo (No. 2), Split (No. 3) and Zagreb (No. 4) suggest the geographical distribution of these coins. None of them are provenienced finds but the first two were preserved in museums that collected archaeological material mainly from the territory of ancient Dalmatia - current central Dalmatia and the Dalmatian islands (AMS) or the vast areas of Bosnia and Hercegovina (ZMS). This indicates mint activity which could have developed somewhere in the Dalmatian region and sometime during the reign of Zeno. The appearance of the lettering E for E in the obverse inscription of group I connects this coinage with Salona where tremisses and solidi in the name of Nepos were minted between 474 and 480. Considering the initial date of these issues, Salonitan tremisses in the name of Zeno could hardly be simultaneous to those of Nepos, 266 zeljko demo (474-480) for a number of stylistic reasons.33 In accordance with contemporary monetary practice in Constantinople, where Zeno did not ignore the name of his co-emperor and son Leo II during the ten months of their joint reign (9 February to November 474),34 it would also be necessary to reject the possibility that Nepos, already entitled patricius while still in Dalmatia, ordered minting in the name of Zeno alone in the short period between 9 February and 24 June 474. Besides these, there is only one other event closely related to the history of Dalmatia and also Salona commemorable in the reign of Zeno. If the coinage reflected the political changes it would not be unusual for the new monetary issue to substantiate the character and delineate the development of historical events in Dalmatia after the end of Nepos' reign. Nepos died on 9 May 480 treacherously murdered by his own comites Ovida (or Odiva) and Viator. The former seized Dalmatia for himself and remained in power for at least another year and a half. It may be assumed that as soon as Nepos was overthrown, Ovida took control over the imperial mint in order to create a coinage suitable to his local requirements.35 The circumstances surround- ing the death of Nepos did not permit him to continue the unified style and aesthetic direction of his predecessor. The new stylistic principles were necessary but when established presumably corresponded to the political motivations of the rebellion itself. The imperial authority of Zeno is clearly emphasized in the first half, while the form PERPT in the second half of the obverse inscription is as original as it is local. The latter is more evident in the style of the attire worn by the emperor {dalmatica?),36 while an original component is represented in the easternized design of the wreath on an essentially western cross-in-wreath reverse. Another line of western influence appears in the ligatured form of /VC while the exergue COMOB denotes an ancestral relationship to the western administrative authority. Considering the provenience, the combination of various stylistic 33 The ligatured form of A VC (=/\l( and A/C) is not relevant at this point in spite of its presence in group I of Nepos' Salonitan tremisses. 34 For specific aspects of the eastern coinage in the period 473-476, see: J. P. C. Kent, 'Zeno and Leo, the Most Noble Caesars', NC XIX6 1959, 93-98, PI. VIIW--4, 35 No solidi of Ovida have been registered to date. However, an irregular solidus without provenience preserved in the AMZ (Inv. No. 279) should be noted at this point. It is seemingly a barbarian issue, minted in the name of Zeno (PI. 2:24). The coin bears an accurate obverse inscription DNZENO PERPFAVC in a form related to the solidi of Odovacar minted in Ravenna not prior to 482. The reverse exhibits Victory with a long jewelled cross, a star (6) in the field right, CONOB in the exergue and the inscription VICTORI AAVCCCS is well engraved with good lettering. The final S could be important for a future attribution of this example. Despite the presence of this letter, this coin unfortunately cannot be assigned to the mint in Salona. CONOB in the exergue even suggests that the origin and minting place should be sought outside the boundaries of praefectura Italiae (ie. neither in Dalmatia nor Italy). 36 On the origin and appearance of dalmatica see: C. Bayet, s. v. »Dalmatica«, Dictionaire des antiquites grecques et romaines d'apres les textes et les monuments (eds. Ch. Daremberg-Edm. Saglio), T. II, Paris 1892, 19-21; A. Mau, s. v. »Dalmatica«, Re IV, 1901, 2025 f.. Several attempts have been made to identify representations of dalmaticae on the stone monuments of Dalmatia, but opinions have varied: I. CremoSnik, 'Nosnja na rimskim spomenicima u Bosni i Hercegovini', GZMS n. s. XVIII, 1963, 104 ff.; D. Basler, 'Kasnoanticko doba. Primjenjena umjetnost', Kulturna istorija Bosne i Hercegovine od najstarijih vremena do turske vladavine, Sarajevo 1966, 360; A. StipCeviC, Iliri: povijest, zivot, kultura, Zagreb 1974, 94. _the mint in salona: nepos and ovida (474-481/2)_267 elements, the motivation of the initiator as well as the significance of the historical evidence, it is not surprising that the Dalmatian non-imperial coinage (tremisses) in the name of Zeno, with its distinctive and autonomous character, ended only when Odoachar devicto Ovidam atque intefecto regnum late proeliis etferro exendit (Auct. Haun., ordo prior, s. a. 482). Key to the Plates pi. l 1 Nepos, solidus, Mediolanum 474: Miinzen u. Medaillen 44/1971, No. 206 2 Nepos, tremissis, Mediolanum 474: Ars Classica XV/1930, No. 1987 [= Cahn A. 24/1912, No. 1932] 3 Nepos, tremissis, Mediolanum 474/475: Hess-Leu 2/4/1958, No. 409 4 Romulus Augustulus, solidus, Mediolanum 475: AMZ No. 25474 5 Olybrius, tremissis, Ravenna 472: J. Sambon, Catalogue de la collection Baxter de Florance, Roma 1887, No. 1913 [= O. U. Bansa, o. c. (p. 250 n. 11), Tav. N:b (Roma)] 6 Glycerius, tremissis, Ravenna 473/474: Catalogue des monnaies d'or. Musee de Saint-Omer, Arras 1983, No. 151 7 Nepos, solidus, Ravenna 474: Ratto R. 13/12/1912, No. 1931 [= O. Voetter, Romische Miinzen und Medaillons. Sammlung Bachofen von Echt, Wien 1903, No. 2801] 8 Nepos, solidus, Ravenna 474/475: Collection du Vicomte de Ponton dAmecourt. Monnaies d'or romaines et byzantines, Paris 1887, No. 827 9 Nepos, tremissis, Ravenna 474/475: Miinzen u. Medaillen XVII/1957, No. 654 10 Nepos, 74 siliqua, Ravenna 474/475: O. U. Bansa, o. c. (p. 250 n. 11), Tav. 0:b 11 Basiliscus, solidus, Ravenna 475: Imperi romano e bizantino regni barbarici in Italia attraverso le monete del Museo Nazionale di Ravenna, Ravenna 1983, No. 87/2254 12 Romulus Augustulus, solidus, Ravenna 475: Collection du Vicomte de Ponton dAme- court. Monnaies d'or romaines et byzantines, Paris 1887, No. 829 [= Hirsch J. XXIV/1909, No. 2856 (coll. Weber) = E. Cocchi-Ercolani, Catalogo della collezione numismatica di Carlo Piancastelli. Monetazione Romana Imperiale 305-476 d. C, Forli 1980, No. 829] 13 Odovacar for Zeno, solidus, Ravenna 480/491: J. P. C. Kent, o. c. (p. 249 n. 8), Taf. XI: 1 [= R. A. G. Carson, Principal Coins of the Romans III: The Dominate, A. D. 294-498, London 1981, No. 1645] 14 Odovacar for Zeno, tremissis, Ravenna 480/491: Kunst u. Miinzen XXII/1981, No. 516 268 zeljko demo PI. 2 15 Nepos, solidus, Roma 474/475 (?): J. Fagerlie, o. c. (p. 260 n. 27), No. 179 16 Nepos, tremissis, Roma 474/475 (?): AMS No. 343 17 Romulus Augustulus, solidus, Roma 475: Leu 22/1979, No. 414 [= G. Lacam, o. c. (p. 248 n. 8), p. 728 Type 5:2, p. 727 and PI. CLXXXII] 18 Nepos, solidus, Arelate 474/475: Hess-Leu 28/1965, No. 546 [= Leu 25/1980, 465 = G. Lacam, o. c. (p. 248 n. 8), p. 601 Type 2, p. 603 and PI. CXLVI] 19 Romulus Augustulus, solidus, Arelate 476? : G. Mazzini, Monete Imperiali Romane, Vol. V, Milano 1958, No. 6 [= O. U. Bansa, o. c. (p. 250 n. 11), Tav. 0:m = G. Lacam, o. c. (p, 248 n. 8), p. 733, p. 735 and PI. CLXXXIV] 20 Nepos, solidus, Ravenna or Roma 476-: Leu-Numismatic Fine Arts 16-18/10/1984, No. 384 21 Nepos, solidus,"uncertain mint/Gallia 476/480: Bourgey 16-18/12/1913, No. 773 22 Nepos, solidus, uncertain mint/Gallia 476 (?): Riechmann XX/1922, No. 1480 23 Leo I (or II), tremissis, an eastern mint 457/475 (?): NMLj 24 Zeno, solidus, barbarian mint under western influence 480/491: AMZ No. 279 25 Theodosius II, nummus, Nicomedia before 450: Aufhauser 3/1986, No. 311 26 Marcianus, nummus, Constantinopolis 450/457: Kricheldorf XXXIV/1980, No. 637 Abbreviations AMS - Arheoloski muzej, Split AMZ - Arheoloski muzej, Zagreb NMLj - Narodni muzej, Numizmaticni kabinet, Ljubljana ZMS - Zemaljski muzej, Sarajevo STUDIA NUMISMATICA LABACENSIA ALEXANDRO JELOCNIK OBLATA Ediderunt Peter Kos - Zeljko Demo LJUBLJANA 1988