^{page 277 note 1} For the Saepinum conductores see C.I.L. 9. 2438Google Scholar of a. 168; cf. the imperial flocks in Egypt, discussed by Johnson, A. C. in An Economic Survey of Ancient Rome, ed. Frank, T. (Baltimore, 1933–1940), ii. 333,Google Scholar or in Africa, C.I.L. 8. 25902 III 17–20. Slaves a iuvencis in C.I.L. 6. 8865 and 8868, and later flocks in Cod. Theod. 9. 30. 2 (364), in a constitution addressed to the consularis of Campania; cf. 7. 7. 1–2. The praepositus camellorum is found in A.E. 1955, no. 181; for overseers of iumenta and greges, C.I.L. 6. 8863–4 5 8. 12640; 14. 2299, manceps gregorum dominorum Augg. from Albano; Hirschfeld, O., Die kaiserlichen Verwaltungsbeamten² (Berlin, 1905), p. 137, n. 6, compared to Cod. Theod. 15. 10. 2 (381).Google Scholar

^{page 277 note 2} Steinhausen, J., Trierer Ztschr. vi (1931) 68 ff.,Google Scholar discussing what is very probably an imperial estate north of Treves, of the fourth century, takes it as a great horse farm, while dismissing another possibility, that it was a hunting park. His identification of the site is not convincing. Rostovtzeff suggests (S.E.H.R.E.² p. 648, n. 92) that a third-century Thracian was ‘probably concerned with a large, perhaps an imperial, estate, where the excellent mounts of the Thracian cohorts and alae were bred’; but this is very flimsy.

^{page 278 note 1} Calder, W. M., C.R. xxiv (1910), 12.Google Scholar The last inscription is dated (ibid.) to ‘the early Empire’, the first to the later second century (M.A.M.A. i. 30). Calder did not use the inscription treated by Robert (see next note) nor did Robert use Calder's inscriptions.Google Scholar

^{page 278 note 2} Robert, L., Hellenica x (1955), 46–62, with references to earlier treatments.Google Scholar

^{page 278 note 3} F.H.G. 4. 145.Google Scholar There is, I think, no reason for attaching the Hermogenian horses to any particular Hermogenes, but it is natural to look for some rich and prominent man with connexions in the East, whose estates there might have been confiscated. Likely candidates are the praef. praet. Hermogenes who died in 373 (Dar.-Sag. s.v. equitium, quoting Gothofredus; but the emperor owned Hermogenian horses already in 371, see Cod. Theod. 15. 10. 1); or the procos. Asiae of c. 282–4 (R-E. s.v. Hermogenes, no. 12); or die mag. equitum Orientis killed in 342. The last is kindly suggested by the Reader to this journal.

^{page 278 note 4} Cod. Theod. 15. 10.Google Scholar 1, to the prefect of Rome, a. 371; 8. 7. 22 (426), with reference to the actuarii equorum currulium in the different regiones of Constantinople; and 8. 7. 21, cornicularii equorum currulium civitatum diversarum, officials of both types appointed through the emperor. For horses furnished to Campanian cities see 15. 10. 2; for others furnished to Caesarea, below, p. 279.

^{page 278 note 5} Claudian cited in Ramsay, W. M., J.H.S. viii (1887), 492,Google Scholar n. 3; Cod. Theod. 6. 4. 19; the horoi in Ramsay, W. M., J.H.S. xxxviii (1918), 135–6. But the reading is not wholly satisfactory. Ramsay dates it to c. 400.Google Scholar

^{page 279 note 1} Lifshitz, B., R.E.G. lxx (1957), 119–23.Google Scholar Date (p. 131), probably sixth/seventh century.

^{page 279 note 2} Die Pfalz unter den Römern (Speier, 1929), i. 62–67. He dates the building of the wall to c. 300, its destruction to c. 350.Google Scholar

^{page 279 note 3} Mouterde, R., Mél. Univ. St-Joseph, xxv (1942–1943), 41–46;Google Scholaribid. xxxiv (1957), 230–3; Pliny, , N.H. 12. 113,Google Scholar on the balsam monopoly in Judaea, seritque nunc eum fiscus. Mouterde identifies the four species as the cypress, cedar, juniper, and fir (‘sapin’). The cypress groves at Daphne in Syria belonged to the res privata (Downey, G., History of Antioch [Princeton, 1961], p. 436, n. 147).CrossRefGoogle Scholar

^{page 279 note 4} On the Fayûvim papyrus marshes see MacMullen, R., Aegpptus xxxviii (1958), 185–6,Google Scholar with references, 186, n. 1; add Wiegand, T., Abh. d. Preuss. Akad. Wissen., 1932, no. 5, p. 46,Google Scholar and Loane, H. J., Class. Phil. xxxix (1944), 10–13.CrossRefGoogle Scholar

^{page 280 note 1} Bloch, H., I bolli laterizi (Roma, 1947),Google Scholar passim, serves to date C.I.L, 8. 22632, p ff.Google Scholar and 22636, 1–2 (dolia). The majority of the stamps belong to the period roughly 90–130. C.I.L. 8. 22632, 3–4 and 30 were found at Utica, a few others at Cherchel, Constantine, and Hadrumetum.

^{page 280 note 2} C.I.L. 5. 8110, 170–4 and 327; 8. 10475, 23a–c; 12. 5678, 8; 13. 10005, 8–9; Labrousse, M., Mél. d'arch. et d'hist. de l'École Fr. de Rome lv (1938), 89 ff.;Google ScholarGrenier, A., Manuel d'arch. gallo-romaine, iii (1958), 80.Google Scholar

^{page 280 note 3} On the Parisian kilns there are many opinions—more recently, Frank, T., Econ. Survey, v (1940), 208;Google ScholarBrusin, G. in Mél. Abramic (Split, 1954–1957), i. 149 ff.;Google Scholar and Panciera, S., Vita economica di Aquileia (Venezia, 1957) 36–38.Google Scholar For occurrences of the stamp outside Italy see (apart from the C.I.L.) I. Petricioli, , Vjesnik za arheologiju i historiju Dalmatinsku liv (1952), 201;Google ScholarMacrea, M., Materiale si cercetari arheologice vii (1958), 378,Google Scholar a Pansian jar from Dacia; and Callender, M. H., Arch. Ael.⁴ xxvii (1949), 65.Google Scholar

^{page 280 note 4} Brusin, G., op. cit. 150–5;Google ScholarPanciera, S., op. cit. 38.Google Scholar

^{page 280 note 5} Gnirs, A., Jb.f. Altertumskunde iv (1910), 79 and 84–88;Google ScholarRostovteeff, M., S.E.H.R.E.² p. 611, n. 26.Google Scholar

^{page 281 note 1} In the last dozen years most interesting studies of western amphora stamps have been made, by M. H. Callender, op. cit., and especially by Thevenot, E., whose latest article (Rev. arch, de l'Est et du Centre-Est x [1959], 320 ff.)Google Scholar contains references to his own earlier work and to the articles of Pelichet, Will, Frank, étienne, Keune, etc. Add Ludlum, C. P., Mem. Am. Acad. Rome xv (1936),Google Scholar and Frank, T., Econ. Survey, v. 82.Google Scholar

^{page 281 note 2} In Baetica lay the towns of Barba, Barbariana, and Barbesula (R.E. s.vv.), perhaps remembering the same firm, or its founder; cf. the modern Juan Barba. The three Augusti are Septimius Severus and his sons (209–11), the stamp recorded in Rome and in die western provinces (e.g. C.I.L. 13. 10002, 1; 15. 2559; Callender, M. H., op. cit. 77).Google Scholar Callender, loc. cit., has pointed out that, once the estates were confiscated, they ceased abruptly to export to Britain, though they continued to find Gallic customers (confirmed by Thevenot, E., op. cit. 222 and 225). The reason, he suggests, is die diversion of more oil to congiaria in the capital.Google Scholar

^{page 281 note 3} Nostrand, Van, Econ. Survey, iii (1937), 185.Google Scholar

^{page 281 note 4} The chief text is P. Fay. 36 (111/12), widi the editors' commentary; Johnson, A. C., Econ. Survey, ii. 330–1;Google Scholar and Reil, T., Beiträge zur Kenntniss des Gewerbes im hellenistischen Aegypten (Bbrna–Leipzig, 1913), p. 17,Google Scholar citing also W.O. 1431, 1433, and 1582, with straw for brick kilns delivered perhaps as a tax to some state factory. The imperial monopoly on manufacture, which actually served only as die basis for a trades tax, does not seem to have extended beyond the Fayûm, nor beyond Trajan's day.

^{page 281 note 5} Thouvenot, R., ‘Les manufactures impériales au Maroc romain’, Publications du Service des Antiquités du Maroc x (1954), 213–16;Google ScholarA.E. 1909, no. 72; B.C.T.H. 1948, pp. 526–7Google Scholar (amphora), and 1954, p. 63; C.I.L. 8. 22632, 1–2; and Callender, M. H., op. cit. 97,Google Scholar doubting Dressel's completion (C.I.L. 15. 2986), f(ig). C(aesareae) p(rovinciae) M(auretaniae).

^{page 281 note 6} A.E. 1903, no. 172; C.I.L. 3. 7419c, d(omini) n(ostri); C.I.L. 7. 1331, 1, from Soudiwark, and Collingwood, , Econ. Survey, iii. 102, the imperial tileries near Silchester, of Nero's time.Google Scholar

^{page 281 note 7} There is the unexplained amphora stamp AAI ii/AUGUST IMP from the lower Rhine (C.I.L. 13. 10003, 119) and the brickyards of Trèves contributing largely to imperial building, which may have been owned by the emperor; but cf. Keune, J. B., Trierer Z. schr. x (1935), 61.Google Scholar Then, too, there are the brick stamps from Bulgaria, A.E. 1944, nos. 11–12, reading AUGG SAR and AUGG AUX, for which die editors offer no explanation. The Upper German potter Sarmus (R.E. s.v.) can hardly have exported to so distant a point, nor are the jars stamped AUX (C.I.L. 8. 10479, 10 and 12. 5686, 1098) of any help. Possibly the bricks were stamped by fourth-century army officers (cf. S. Soproni, , Arch. Ertesitö lxxxv [1958], 52 ff.;Google Scholar and C.I.L. 3. 3761 ff.Google Scholar and 4668; 10677 ff., 11376, 11856, and vol. 3, p. 2328, 197), but they remain obscure.

^{page 282 note 1} On constructions per colonos, see C.I.L. 8. 8701–2, 8777, 8828;Google Scholar 9. 2828. For the straw for brick-kilns, C.I.L. 8. 14428, line 8, and above, p. 281, n. 4; and for stamps of Mursa and Augusta Vindelicum, C.I.L. 3, P. 711;Google Scholar 3. 3774, 1–5; 10694, 1–4; and 14596. The legate's bricks, and the Vespasianic baths, are to be found in A.E. 1908, no. 7— sub Augusto fere, P.I.R.¹ s.v. Petrucidius— and I.G.R.R. 3. 659.

^{page 282 note 2} So the well-known protocol on Spanish amphorae, fisci rationis patrimonii (above, p. 281); perhaps also the amphorae stamped IMPE VECT? A //so; (Vect … for vectigal?) of C.I.L. 5. 8112, 5, and the stamp ATRIM = ? p]ATRIM(ONII) ? of C.I.L. 3. 14373. 1; cf. vectigal patrimonii on glass bottles (A.E. 1914, no. 292) and PATRIMONI on others (MacMullen, R., op. cit. 185).Google Scholar