¹¹² Inscriptions latines de l'Algérie 11.2, ed. H.-G. Pflaum post S. Gsell (1976), No. 6180 (= ILAlg).

¹¹³ This alternative occurred to me when Eric Birley drew my attention to Pflaum's reading.

¹¹⁴ In the Notitia Dignitatum, the ala V Praelectorum appears in Egypt as the garrison of Dionysias (Or. 28,34) also P. Abinn. 59,12 (A.D. 345), while the ala VII Valeria Praelectorum appears in Osrhoene (Or. 35,27).

¹¹⁵ As Dr Walter Cockle points out to me, citing Apuleius, Met. 7, 11. I have to thank Dr Cockle for much help with lexicography.

¹¹⁶ BGU 696 (= R.O. Fink, Roman Military Records on Papyrus (1971), No. 64; Loeb, Select Papyri ii, 401) i, 25-30.

¹¹⁷ The precise meaning of vircam (correctly virgam) in this context is rather obscure, but it is hardly in doubt that it broadly refers to the authority of a decurion (at least in a cohors equitata) in exactly the same way as vitis refers to the authority of a legionary centurion.

¹¹⁸ For the correct form Bravoniacum see A.L.F. Rivet and C. Smith, The Place-Names of Roman Britain (1979), s.v.

¹¹⁹ Dobson, B. and Mann, J.C., Britannia iv (1973), 191–206.CrossRefGoogle Scholar

¹²⁰ J.C. Mann, Legionary Recruitment and Veteran Settlement during the Principate (1983), tables 1–4.

¹²¹ III Augusta early declared for Severus, hence its receipt of the titles pia vindex.

¹²² Herodian, for what he is worth, refers to Severus ‘drawing troops from all sides’ and comments on the size of the force he collected (III. 14. 3–4). CIL VI. 2464 = ILS 2089 with CIL VI. 1057 indicates the presence of men from the Praetorian Guard, while CIL XIII. 3496 = ILS 9123 and RIB 2216 may suggest the presence of a detachment of XXII Primigenia from Upper Germany. A detachment from III Augusta would hardly be out of place.

¹²³ JRS lvii (1967), 205 No. 16 and pl. xix, 1Google Scholar: the reference to the men from the two Germanies is difficult to decipher on the stone. It either reads EXER, for EXER(citui), or EXPL, for EXPL(oratorum). In either case, there is a strong suggestion that these men have been brought over to fight, not merely (e.g.) for construction work, although reinforcement in the aftermath of disaffection is possible, as suggested by Birley, E., Epigraphische Studien 4 (1967), 103–7Google Scholar. More recently a further inscription has emerged at Piercebridge referring to a man from Germania Superior: Britannia xvii (1986), 438, No. 20 and fig. 7.Google Scholar

¹²⁴ Review of RIB in Bonner Jahrbücher clxvi (1966), 642–3.Google Scholar

¹²⁵ Here reverting to Collingwood's reading DOM[u in preference to Alföldy's DON[avit.

¹²⁶ There are of course in the frontier area a large number of undated inscriptions set up by men described simply as Germani (RIB 920, 1525 and 2063), or with German names (RIB 926), or as specifically coming from Upper Germany (RIB 1620 and 1665). Conceivably one or more of these inscriptions may date to c. A.D. 217.

¹²⁷ Mann, op. cit. (note 120), 18. (It is not very likely, however, that Agathopus began his service in III Augusta.) This recruitment of citizens to the alae and cohorts of Numidia may have some connection with the absence of military diplomas issued to men of that army.

¹²⁸ Acta. Cons. Carth., A.D. 256. Unfortunately, Agathopus' use of patria does not quite clinch city status, nor do the sacerdotes who appear in ILAlg 6176 and 6178; for what it is worth, Procopius (Bell. Vand. 2.15) refers to the territory of Gadiaufala.

¹²⁹ ILAlg 6179.