¹ ZfN xv (1887), 239–50.

² Abh. d. Königl. Akad. der Wissensch., Berlin (1862), 489–531.

³ CRAI (1908), 254–66.

⁴ Seston, W., Dioclétien et la Tétrarchie (1946), 339.Google Scholar

⁵ Cf. Callu, J.-P., La politique monétaire des empereurs romains de 238 à 311 (1969), 389, n. 2.Google Scholar

⁶ Sutherland, C. H. V., Roman Imperial Coinage VI (1967), 88–93.Google Scholar

⁷ Bruun, P. M., RIC VII (1966), 13–24Google Scholar; also Eranos 1962, 93–100.

⁸ Univ. of Birmingham Hist. Journ. XII (2), 1970, 129–54. These two articles should be read in conjunction, for each complements the other. For the intervening period, 324—c. 400, see: Numismatic Chronicle VIII² (1972, forthcoming).

⁹ Jones, A. H. M., JRS XLIV (1954), 21–9Google Scholar; Chastagnol, A., La Prefecture Urbaine à Rome sous le Bas-Empire (1960), 3–4.Google Scholar

¹⁰ Jones, A. H. M., Later Roman Empire (1964) III, 17Google Scholar (n. 66)—citing CTh. XI, 3, 2 (327).

¹¹ Jones, A. H. M., JThS NS, v (1954), 224–7.Google Scholar

¹² It had long acted as an independent unit for the purposes of both res summa and res privata: see p. 79 below.

¹³ Jones, LRE III, 4 (n. 16); cf. n. 34 below. Use of the term ‘Suburbicaria’ is parachronistic but convenient.

¹⁴ For the mints: Sutherland, RIC VI, 5–6; Bruun, RIC VII, under mint headings. For the dioceses: Laterculus Veronensis (ed. Seeck, , Notitia Dignitatum (1876), 247–51).Google Scholar A Constantinian pro(curator) s(acrae) m(onetae) u(rbis). occurs in CIL VI, 1145. Not. Dig. Occ. XI, 38–44 represents the modified Diocletianic situation. Monetae tended to be accompanied by thesauri, under praepositi, for obvious reasons of convenience (Occ. XI, 21–37). cf. Jones, , LRE III, 104–5 (n. 44), 112–3 (n. 62).Google Scholar

¹⁵ The much reduced province of Africa Proconsularis in which Carthage itself stood was of course not subject to the vicarius Africae (nor to the praetorian prefects for that matter) but to its proconsul, the vicar being, in theory at least, forbidden entry (e.g. CTh. 1, 15, 10). This would not inhibit the use of its mint by both and, since many of the laws in CTh. addressed to the vicar were (where recorded) accepted or posted at Carthage, he must in fact have been as frequently resident there. The administrative boundaries of both res summa and res privata remained unaffected by this anomaly: see below and, for the position of proconsuls, Jones LRE III, 5 (n. 17). The situation seems to have had much in common with that existing between the praefectus and the vicarius urbis (Romae).

¹⁶ Tarraco in Mommsen's list. The attribution to Ticinum is now universally accepted: cf. Sutherland, RIC VI, 6–7.

¹⁷ Carausius and Allectus: Carson, R. A. G., Journ. Brit. Arch. Assoc. (1959), 33–40CrossRefGoogle Scholar; id., Mints, Dies and Currency (1971), 57–65. Diocletian: Sutherland, RIC VI, 113, 115.

¹⁸ Sutherland, RIC VI, 645–7: which account is to be preferred to those of Schwarz and Callu arguing for 296 (cf. Callu, op. cit. 190–3, 386–7).

¹⁹ Webb, P. H., RIC V(I) (1927), 256–62.Google Scholar

²⁰ Sutherland, RIC VI, 486, 501, 505. A brief issue of gold coin was also struck at Serdica by Licinius in 313–4: Bruun, RIC VII, 478.

²¹ According to Aurelius Victor (Lib. de Caes. 40, 17) Alexander was ‘apud Poenos pro praefecto gerens’. He was thus vicarius Africae. This is confirmed by Zosimus (Hist. II, 12), who adds that Maxentius had demanded that Alexander send his son to him as hostage, for fear that he would otherwise revolt. The closure of the mint of Carthage is further evidence of this attitude. For the reign of Alexander see, in the last instance, R. Andreotti in Afrika und Rom in der Antike (edd. H.-J. Diesner, H. Barth, H.-D. Zimmerman, 1968), 245–76. Sutherland (RIC VI, 411) emphasizes the extraordinary nature of the mint of Carthage. Certainly it was set up either shortly before or shortly after Maximian's arrival on campaign against the Quinquegentiani (297–8), and it consistently used variant designs on its products. But its continuance until 307 argues the fulfilment of more permanent needs. Seston (op. cit. 116–7) is sceptical of the military necessity for Maximian's visit, but (325–31) supposes several of the administrative rearrangements of the period to have resulted from it. This is not incompatible with what has been suggested of the mint.

²² Carthage-Ostia: Sutherland, RIC VI, 393–4. Ostia-Arles: Bruun, RIC VII, 227, and The Constantinian Coinage of Atelate (1953), 5–16.

²³ Bruun, RIC VII, 462–6; cf. n. 58 below.

²⁴ The production of gold and silver was never continuous (cf. n. 58 below), but even that of copper was occasionally interrupted for reasons that are not always now apparent: Sutherland, RIC VI, 37–73; Bruun, RIC VII, 87 (for copper only). It is over the longer term that the continuity of these mints becomes apparent.

²⁵ Carthage and Aries being treated as one mint.

²⁶ The successful campaigns of Constantius. See n. 17 above.

²⁷ It is immaterial to the present treatment whether or not the Caesars were created simultaneously; cf. Jones, LRE III, 3 (n. 4).

²⁸ Webb, P. H., RIC v(II) (1933), 218.Google Scholar It would not be surprising if there were some connection between the closure of Tripolis and the opening of the mint of Heraclea.

²⁹ The ancient—and for the most part mediaeval—world lacked monetary policies: monetary events did not anticipate but reacted to situations. See: Jones, A. H. M. in Essays in Roman Coinage Presented to Harold Mattingly (1956), 26Google Scholar; Crawford, M. H., JRS LX (1970), 40–8.Google Scholar

³⁰ Seston, op. cit. 337.

³¹ Acta Marcelli, Anal. Boll. XLI (1923), 262. The equation of agens vices with vicarius was established in its essentials by E. Michon in an article still useful as a collection of sources: Mém. Soc. Nat. Ant. de France LXXIV (1914), 244–99. The latest treatment of the question—by M. T. W. Arnheim, who seems unaware of Michon's article—merely reaches the same conclusion: Historia XIX (1970), 593–606.

³² See n. 40 below.

³³ ILS 619. For a list of early vicars see Jones, LRE III, 5 (n. 17).

³⁴ Seston, op. cit. 337, n. 4. See also Chastagnol, op. cit. 26 (n. 1)—to whose account of the vicarius urbis (or vicarius in urbe Roma) and vicarius praefecturae urbis that of Jones (LRE III, 4–5 (n. 16)) is to be preferred.

³⁵ Seston's proposed dating was 297/98 (op. cit. 333). But T. Flavius Postumius Titianus can be shown to have held the posts of both corr. Campaniae and corr. Italiae reg. Transpadanae before his ordinary consulship in 301 and his proconsulship of Africa in 295. See Chastagnol, op. cit. 21–5, where a date of 290/91 is suggested.

³⁶ AE 1942–3, 81. Alexander was still vicar under Maxentius (edd. Reynolds, J. M., Ward-Perkins, J. B., The Inscriptions of Roman Tripolitania (1952), 464)Google Scholar. In view of the information in n. 21 above, it seems worthwhile asking whether Valerius and Domitius Alexander might not be the same person. On this see Jones, A. H. M., Martindale, J. R., Morris, J., The Prosopography of the Later Roman Empire I (1971). PP. 43–4. nos. 17, 20.Google Scholar

³⁷ Seston, op. cit. 339, n. 3.

³⁸ Lactantius, De Mort. Pers. VII, 4.

³⁹ For a list of early rationales see Jones, LRE III, 5 (n. 18), and 104 (n. 43); for early magistri ibid. 5 (n. 19), 95–6 (nn. 1, 2); also ibid. 344–6 (Appendix I). cf. n. 49, below.

⁴⁰ P. Beatty Panop. I (Sept. 298), 2 (Jan.—Mar. 300).

⁴¹ Vandersleyen, C., Chronologie des preféts d'Égypte de 284 à 395 (1962; Coll. Lat. 55), 110–4.Google Scholar

⁴² Jones, LRE III, 16 (n. 61). The status of Aemilius Rusticianus ὁ διασημ(ότατος) διαδεχό(μενος) τὰ μέρη τῶν ἐξοχωτάτων ἐπάρχων of P.Oxy. 1469 (298) remains uncertain. He was just possibly an early vicarius Orientis.

⁴³ To n. 40 above, add Jones, LRE III, 104 (n. 41).

⁴⁴ Vandersleyen, op. cit. 113–4; ed. Skeat, T. C., Papyri from Panopolis (1964), xviii, 153–6 (P. Lond. 1260).Google Scholar

⁴⁵ See n. 30 above.

⁴⁶ Jones, LRE III, 5 (n. 22), 6 (n. 24).

⁴⁷ Constantius possessed a defined territorial jurisdiction: it remains uncertain whether Galerius did. The documentary evidence seems negative, although his extensive building operations in Thessalonica might be taken as implying the opposite. The coinage argues for the kind of administrative distinction between Constantius and Maximian that may have derived from the former's possession of his own chief rationalis as well as his prefect; cf. Jones, LRE III, 3 (n. 7), 6 (n. 24).

⁴⁸ Sutherland, RIC VI, 229; Bruun, RIC VII, 121. Curiously, and probably significantly, the same tended to happen later at Alexandria which was formally in a comparable administrative situation: RIC VII, 698; Pearce, J. W. E., RIC IX (London 1951), 296.Google Scholar

⁴⁹ C.s.I.—Not. Dig. Or. XIII, 5; Occ. XI, 10–20. C.r.p.—Not. Dig. Or. XIV, 4; Occ. XII, 6–15. Jones (LRE 1, 19) supposes the rationalis r. p. per Siciliam of Occ. XII, 10 to have been a late addition. The early division of the suburbicarian diocese for the purposes of the res summa is confirmed by CTh. XII, 6, 2; XII, 7, 1 (both 325 and addressed Ad Eufrasium rationalem trium provinciarum), and by XI, 30, 14 (327, Victori rationali urb. Rom.); cf. n. 39 above.

⁵⁰ Not. Dig. Occ. XI, 13.

⁵¹ Not. Dig. Occ. XI, 14. For hoard figures see Callu, op. cit. 455; Salama, P., Libya Antiqua III–IV (1966–1967), 21–7.Google Scholar

⁵² Stein, E., Studien zur Geschichte des byzantinischen Reiches (1919), 144–7Google Scholar; Kent, J. P. C. in Cruikshank Dodd, E., Byzantine Silver Stamps (1961), 36–7.Google Scholar

⁵³ Although it is noticeable that the provisions of a procedural reform of Valentinian and Valens, resulting in the minting of precious metal being confined to the comitatensian largitiones, seem subsequently to have been relaxed to allow minting at prefectural headquarters—even if it was performed only by moneyers seconded for the purpose. See J. P. C. Kent in Essays in Roman Coinage Presented to Harold Mattingly, 199–203; Hendy, op. cit. 142–3.

⁵⁴ Lactantius, De Mort. Pers. VII, 9, describing Diocletian's ‘infinita quaedam cupiditas aedificandi’: ‘Hic basilicae, hic circus, hic moneta, hic armorum fabrica, hic uxori domus, hic filiae.’

⁵⁵ Jones, in Essays in Roman Coinage Presented to Harold Mattingly, 15–6. Not invalidated (in particular for the late Roman period) by Sutherland, C. H. V. in JRS XLIX (1959), 46–55, and LIII (1963), 14–20.Google Scholar Propaganda on Roman coinage can, at best, never have been more than a secondary consideration. To the extent that the production of coinage was not primarily geared to the needs of the general public it was equally not so for its propagandizing. It was precisely those denominations that were least liable to (socially) wide, rapid circulation (i.e. gold and silver) that had the most rapidly changing design.

⁵⁶ Lactantius De Mort. Pers. VIII, 3: ‘opulentissimae provinciae, vel Africa vel Hispania’.

⁵⁷ The military establishment in Africa was considerably increased at this period, probably as a result of Maximian's visit. But it could, even so, not compare with that of the Rhine, Danube or East. And it should be pointed out that Africa had, in fact, originally been provided with a mint—its closure being due to political reasons (cf. n. 21 above). The establishments in Spain and Viennensis seem to have been minimal. D. van Berchem, L'armée de Dioclétien et la réforme constantinienne 37–49 (Africa); Jones, LRE in, 374 (table IX, Africa), 379 (table XIV Africa), 377 (table XI, Spain and Tingitania). A large proportion of the African military seem to have been tribal limitanei (Jones, LRE III, 201, nn. 103–4, 106). For the Diocletianic calculation (and presumably payment) of military stipendia and donativa etc. in denarii (i.e. copper coin) see P. Beatty Panop. 2. For commentaries: ed. Skeat xxvi-xxx; Jones, LRE III, 187–9 (n. 31). For a Constantinian example: P. Oxy. 1047.

⁵⁸ On the developed theory of the comitatensian (and therefore ‘travelling’) mint for the production of precious-metal coinage at this period, see: Bruun, P. M., Studies in Constantinian Chronology (1961), 23–77Google Scholar, and RIC VII, 13–8; Alföldi, M. R., Die constantinische Goldprägung (1963), 12–20Google Scholar. For a more sceptical view: P. Bastien's review of RIC VII in JRS LVIII (1968), 281; and, for Diocletian, Sutherland, RIC VI, 54–5. While a strong tendency towards a comitatensian pattern is undeniable, even at this stage, its elevation into a virtually exclusive rule is unwarranted—and is moreover disproved by fourth-century accessional and quinquennial issues. It was only with the measures described in n. 53 above that the regular existence of comitatensian mints such as Milan and Ravenna, which were both outside the earlier structure and which produced coinages that were virtually confined to the precious metals, was rendered possible.

⁵⁹ Of the new mints Trier was the main residence of Constantius, Nicomedia that of Diocletian, and Thessalonica probably that of Galerius. The presence of a ruler at these mint cities doubtless influenced even their production of copper coin, but that it was not the predominant influence is seen by their continuing in production not only in the absence—often prolonged—of the ruler, but also into periods when they were no longer residences.

⁶⁰ Hendy, op. cit. 147–52.

⁶¹ Stein, Studien 149–50.