² For the site see Britannia xv (1984), 310. Excavations for the Department of Greater London Archaeology, Museum of London, were directed by Brian Yule. We are deeply indebted to Mr Yule for stimulating discussions and for making available to us his meticulous study of the fragments and their physical relation to one another, as well as to Mike Hammerson for his acute observations on the text at a critical point.Google Scholar

³ Note, however, the non-military use of the word at the Horrea Galbana at Rome, where the personnel were divided into at least three ‘cohorts’: see de Ruggiero Dizionario Epigrafico II 338, where it is suggested that these were enclosed sections of the building – ‘courts’ (Italian corte).

⁴ cf. the very similar though fragmentary inscription from Mainz dedicated by the primi ordines (et centuriones) of legion XXII Primigenia in a.d. 204, CIL xiii 6801.

⁵ As men on detachment from legion III Cyrenaica and XXII Deiotariana were listed on the well-known inscription from Coptos, Egypt, ILS 2483.

⁶ On this inscription see also Rankov, N. B. ‘A contribution to the military and administrative history of Montana’ in Poulter, A. G. (ed.) Ancient Bulgaria (Nottingham 1983), 40–61, especially pp. 52–4. This inscription is No. 265 in Saxer's study of Vexillations in Epigraphische Studien. Cf. also Saxer Nos. 320–323 from Castellum Dimmidi in Algeria, listing men of Legion III Augusta on detachment from Lambaesis, all apparently identified by cohort only.Google Scholar

⁷ In the late empire seconded legionaries on the governor's staff were, perhaps significantly, known as cohortalini, see Jones, A. H. M., ‘The Roman Civil Service (Clerical and Sub-Clerical Grades)’, JRS xxxix (1949), 38–55, esp. p. 49 ffGoogle Scholar. Much the largest group of seconded legionaries in the officium of the legate were beneficiarii consularis, probably 60 in all, see Jones, op. cit. p. 44Google Scholar, and it is just conceivable that this is a list of members of a guild confined to men of this rank, such as is apparently attested at York, the capital of Britannia Inferior in the early third century (Britannia i (1970), 307, No. 12).Google Scholar

⁸ Aurelii: Frag. (b), line 2, Frags. (c)–(g), lines 3, 16, 17 and probably to be restored in Frag. (b), line 5 and (c)–(g) lines 5 and 12. In Frag. (c) 1, where the stem of the name ends in D and there are an estimated five letters missing, the nomen may have been Secund(ius) or Iucund(ius). In Frag. (c)–(g) 13 and 14, the nomina could have been either Iul(ius) or Ael(ius). For Aemilius used as a cognomen, see A. Mocsy, Nomenclator, listing 17 examples of which nine come from Spain. Gontius (line 3) does not seem to be precisely matched.

⁹ During excavations for the Bath Archaeological Trust directed by Professor B. W. Cunliffe. See Britannia xi (1980), 387–8.Google Scholar

¹⁰ This is the only defixio from the deposit to be inscribed on a pewter vessel, although Britannia xiv (1983), 339, No. 5Google Scholar, seems to have been inscribed on a fragment of one. The text is a list of personal names like others from the deposit, in format closest to Britannia xiii (1982), 400, No. 4Google Scholar, where names are glossed maritus, ussor, etc. Like these texts, RIB 154 and Britannia xiii (1982), 404–5, No. 7, in particular, it is a list of personal enemies or persons suspected of theft. The names are the usual mixture of ‘Roman’ (7) and ‘Celtic’ (7), the ‘Roman’ names, four of them elaborations of simpler cognomina, all being typical of Roman Britain. The handwriting is bold and assured: the initial letters of Severianus, Marinianus, Aeternus above all, and Bellaus, are more ‘calligraphic’ than any other cursive text from the deposit.Google Scholar

¹¹ By King Alfred's College, Winchester, directed by Ms A. M. Robinson on behalf of the Winchester Archaeology Office. Information from A. M. French who also supplied a drawing by A. Turner.

¹² cf. RIB 2222 for a milestone of Gordian III (P. F. Aug) also erected by the res publica Belgarum.

¹³ Directed by Michael Audouy of the Northants. Archaeological Unit for the Brixworth Archaeological Research Committee. Full information and photograph from David Parsons.

¹⁴ By the Central Excavation Unit of the H.B.M.C.E., directed by David Neal who supplied full details and a drawing.

¹⁵ The first line of the inscription is set very close to the upper edge of the stone and, since there is no moulding, the text may be the end of a longer inscription which started on a second stone immediately above it. This is perhaps supported by 1. 1 where QV]OD suggests itself. If, however, there was no second stone IN H(ONOREM)] D(OMUS) D(IVINAE) would be a possible restoration; 1. 2 probably HO]NORAT[VS, either as a name or a past participle; 1. 3 there appears to be a gap following the surviving letters, but since the edge of the stone is broken diagonally it would be possible to read …]ISCHOL. After the gap on the edge of the break are traces of what could be the left-hand side of the letter A, in which case it would be tempting to read SCHOLA[M, despite the lack of a gap between the s and the first surviving letter of the line. 1. 4, after the s is part of the top of a letter, either an upright or the left-hand end of the cross bar of a T. If this is a personal name, the only one to be suggested in I. Marriott's Index of Personal Names is Venustus.

¹⁶ There are also traces of a roughly scored vertical line to the right of the upright of the D in 1. 1, presumably made by the stone-cutter when marking out the inscription.

¹⁷ Found by the Monmouthshire Antiquarian Association. Information and squeeze from R. J. Brewer of the National Museum of Wales. In 1. 1 FES could be part of the deceased's nomen as Festinius (RIB 587), but is much more likely to have been part of the common cognomen Festus, unusually, for a cognomen, preceding the voting tribe (Galeria) recorded in 1. 2. Spacing also suggests that other details of the deceased were similarly given before the cognomen. For mention of the cognomen before the voting tribe cf. RIB 508 and 525 and for other details given in this position cf. RIB 539.

¹⁸ By the Clwyd-Powys Archaeological Trust directed by Kevin Blockley who supplied details, a photograph and a squeeze.

¹⁹ For the site see Britannia viii (1979), 358–9. The bath-house is the southernmost of the two structures labelled ‘Excavated Roman Buildings’ on the plan, FIG. 3.Google Scholar

²⁰ The final letter could be a D, E, F, or P.

²¹ By the Excavation Section of the Grosvenor Museum, directed by T. J. Strickland. Information and photograph from P. Carrington of the Grosvenor Museum.

²² The road is that running between the barracks marked ‘C’ and the building marked ‘D’ in Britannia xiv (1983), 298Google Scholar, FIG. 8. For the site see Britannia xiii (1982), 352.Google Scholar

²³ The N is extremely attenuated.

²⁴ By the Department of the Environment directed by D. F. Petch. Information from D. J. P. Mason. For the site see JRS liv (1964), 156.Google Scholar

²⁵ For the omission of a number after mentioning the grant of tribunician power, followed immediately by the title p(ater) p(atriae) see RIB 330, 397, 398 - all in the titulature of Trajan.

²⁶ By a person unknown, who through an intermediary allowed Mr P. T. Bidwell to photograph it and make a squeeze.

²⁷ The fragment seems too slight for restoration. ]PL[ suggests templum or a personal name (simplex, placidus, etc., and cognates), ]ER per, fecer(unt), frater, mater, pater, etc., and CO[ cohors, coniunx, etc. ]ENNIO is probably part of a personal name in the dative or ablative case (Herennio, Ennio, etc.). If ]VET[ begins a word, veteranus or vetustate are possibilities, the trace of the next letter favouring a V. If the stone is not a gravestone, at a guess it records the restoration of a temple by a cohort (per cohortem) commanded by an officer (instante […]ennio); but the vetustate conlapsum formula would be oddly delayed.

²⁸ During excavation for the Tyne and Wear Museums Service directed by Messrs P. T. Bidwell and R. Miket. Mr Bidwell sent a squeeze and full details. The stone is now in the site museum, where RSOT has seen it.

²⁹ See RIB 1202 with note ad loc., and Britannia xi (1980), 405, No. 6Google Scholar. Mr Bidwell points out that this is the earliest epigraphic evidence of the presence of Coh. V Gallorum at South Shields (cf. RIB 1060, a.d. 222), only two years after the campaigns of Septimius Severus, for which the fort was converted into a supply-base. Leaden sealings of the cohort (CVG) found at South Shields have usually been dated to the supply-base period. Despite the reasonable scepticism of Dore, J. N. and Gillam, J. P., The Roman Fort at South Shields (1979), 67–8Google Scholar, it now looks as if the cohort was already there in 209-11, even if its main body was elsewhere, Cramond for instance (cf. RIB 2134). Since the principia was reinstated in c. 220 and this stone re-used, the demonstration of loyalty to Caracalla and his mother was probably scrapped when news came of their deaths in 217.

³⁰ The restoration is based upon the similar dedication (but apparently omitting communi) found in 1979 at Newcastle (Britannia xi (1980), 405, No. 6Google Scholar, with notes ad loc.). The width of line is known from the certain restoration of lines 3 and 4 as […]dev [otione] | [com] muni[…]. The restoration remains approximate, since it is uncertain how words were abbreviated or even divided between lines, in view of the variables introduced by ligatures (line 1), letters crowded together (line 4), and the different possibilities of abbreviation or division.

³¹ By Mr M. Finlay, who sold it in 1985 to the Cumbria Park Hotel, Stanwix, where it now is. The original provenance is unknown, but Birdoswald fort is less than two miles away (cf. RIB 1899, Mars and Victory).

³² Mr I. Caruana, who is publishing the altar in CW² lxxxvi (1986), forthcoming, sent a drawing and a draft of his paper. He points out that the inscription is anomalous, in that the dedication is in the nominative case and the text runs from the side to the front (cf. RIB 1366, with note ad loc.), but it seems to be genuine.

³³ L. J. F. Keppie, ‘Roman inscriptions from Scotland: some additions and corrections to RIB I ’, PSAS cxiii (1983), 391–404. See II.3 (p. 399). It is now in the National Museum of Antiquities, Edinburgh.Google Scholar

³⁴ ibid., II.I with Fig. 10 (pp. 398–9). It was presented to the Hunterian Museum, Glasgow, before 1897.

³⁵ During the excavations reported in PSAS xxxii (1898), 466. Keppie, op. cit. (note 33), II.2 (p. 398). It is now in the National Museum of Antiquities, Edinburgh. We have noted here and in Addenda and Corrigenda only the most important of Dr Keppie's unpublished contributions; his whole paper is worth reading.Google Scholar

³⁶ Directed by C. J. Drage for the Trent Valley Archaeological Research Committee (now the Trent and Peak Archaeological Trust) supported by the H.B.M.C.E.. Full information and drawing from Graeme Gilbert of the Trust, who submitted the object for inspection.

³⁷ For two altars found nearby, one inscribed, see Britannia xi (1980), 404, No. 3.Google Scholar

³⁸ Directed by J. P. Green for the Dorchester Excavation Committee. Information and drawing (by R. Soames) from Jo Draper.

³⁹ By R. A. Hattatt, who provided full information and drawing. For a very similar brooch with identical inscription, see Britannia xi (1980), 413, No. 45, and note 64 for continental parallels, and No. 40 below. MWCHGoogle Scholar

⁴⁰ Excavations directed by D. Rudling of the Sussex Archaeological Field Unit for East Sussex County Council, the Department of the Environment and Keymer Tiles. Mr Rudling provided a drawing (by S. Foster) and full information.

⁴¹ Compare the tile graffito from Heybridge, Essex, Britannia xiii (1982), 411–12, No. 19 and note 34.Google Scholar

⁴² Directed by John Hope for the Brain Valley Archaeological Society who supplied full information and submitted the object for inspection. For a full discussion see M. Henig above, p. 241.

⁴³ By the Alice Holt survey group, directed by Mr A. B. Lyne, who submitted this and the following item for inspection and provided full information.

⁴⁴ Directed by G. R. Burleigh, who provided drawings by G. Denning and information on this and the following item. For an earlier example from the same site, see Britannia xii (1981), 385–6, No. 40 and below under Corrigenda.Google Scholar

⁴⁵ Identified by X-ray fluorescence as a brass copper-zinc alloy with a little tin and negligible lead content.

⁴⁶ Excavation directed by Paul Bennett for the Canterbury Archaeological Trust. Information from Jane Porter to whom we are indebted for the technical description of the object. Ralph Jackson draws our attention to a similar container identified as an ink-pot from Salzburg, see E. Künzl, ‘Medizinische Instrumente aus Sepulkralfunden der römischen Kaiserzeit’, Bonner Jahrbücher clxxxii (1982), 115, fig. 90.6.

⁴⁷ Information and sketch from Paul Montgomery. Purchaser and present location unknown.

⁴⁸ For a recent discussion of the products of P. Cipius Polybius by J. Bennett and R. Young, see Britannia xii (1981), 37–44.Google Scholar

⁴⁹ By the Trust for Lincolnshire Archaeology. Rubbings and contact drawing by the director of the Trust, M. J. Jones, who provided full information.

⁵⁰ This could conceivably represent parts of two names as SE]VER[VS | DE]CIM[VS.

⁵¹ See the full discussion by Ralph Jackson, Britannia xv (1984), 246–50, fig. 5.Google Scholar

⁵² Reported by Christine Jones of the Museum of London and subsequently purchased by the Museum. Miss Jones also drew our attention to, and provided information on, the following four items.

⁵³ Eve Harris points out that the inscriptions misce or misce me/mi are commonly found on Rhenish motto beakers (CIL xiii 10018, 119–28Google Scholar) where the idea of ‘mixing’, appropriate in the context of diluting wine, also has a sexual connotation; its use on a ring is somewhat unusual but can be paralleled by the inscription on a fibula from Belgium (CIL xiii 10027, 162).Google Scholar

⁵⁴ By Guy de la Bedoyere who supplied a drawing and submitted the object for inspection.

⁵⁵ These are the only names to be suggested by A. Mócsy's Nomenclator.

⁵⁶ We are grateful to Professor Emilio Rodriguez-Almeida for confirming the reading from a photograph.

⁵⁷ Information and photograph from John Schofield.

⁵⁸ By Martin Henig, to whom we are indebted for providing details on the iconography. The aplustre perhaps suggests a naval victory or at least activity during the Severan period. Britannia is discussed by Toynbee, J. M. C., The Hadrianic School. A Chapter in the History of Greek Art (Cambridge 1934), 53–65Google Scholar. The representation follows coin types of Antoninus Pius (pp. 58–60 pl xi, 24, 25) and a medallion of Commodus (pp. 62–3, pl. xii, 1) where Britannia holds a standard rather than an aplustre, and there is no stag, for which see Wright, R. P., ‘A new leaden sealing from York and further examples from Brough-under-Stainmore’, CW N.S. liv (1954), 102–4, fig.IC from Combe Down.Google Scholar

⁵⁹ For the legend cf. RIB 643, York, a dedication by a freedman of two emperors, perhaps Severus and Caracalla.

⁶⁰ For a full discussion by M. Henig, see Ant. Journ. lx (1980), pt. ii, 331–2, with pl. LXB. He suggests that the two emperors are Severus and Caracalla, which is possible on iconographic grounds.Google Scholar

⁶¹ Information from R. A. Hattatt, who kindly provided a drawing.

⁶² By Mr J. J. J. Embleton. Mrs E. Hartley of the Yorkshire Museum sent a drawing and other details.

⁶³ This seems to be the only example known. Since leaden sealings often bear the abbreviated name of a legion, ala or cohort, the obverse should perhaps be expanded as eq(uites) S(…), e.g. the equites Sarmatae of Ribchester (RIB 583, etc.).

⁶⁴ Directed by A. S. Anderson, who with Mike Stone provided information on this and the following items and submitted the objects for inspection. For the site see A. S. Anderson and J. S. Wacher, ‘Excavations at Wanborough, Wiltshire: An Interim Report’, Britannia xi (1980), 115–26.Google Scholar

⁶⁵ Similar to Gillam Form 266 but without the groove on the flange. The vessel has a fine white fabric with white grits.

⁶⁶ Information and drawing from R. A. Hattatt.

⁶⁷ cf. No. 18 above.

⁶⁸ For a full discussion of this important find and the significance of the inscriptions, see Boon, G. C., ‘A Trulleus from Caerleon with a stamp of the First Cavalry Regiments of Thracians’, Ant. Journ. lxiv (1984), pt. ii. 403–7 with fig. 3 and pl. lv.Google Scholar

⁶⁹ Lot 195 in Sotheby's sale of antiquities (etc.), 16 July 1968. We are grateful to Mr Wallace for getting in touch with us over his purchase on the suggestion of Nigel Seeley, MWCH.

⁷⁰ BY RPW. The stone has been in the Museum of Antiquities, Newcastle upon Tyne, since 1963 (JRS liv (1964), 185, item b)Google Scholar. In 1. 1 the first letter is P, not R, as Bosanquet noted (AA³ xviii (1921), 117). The stroke which might seem to be the tail of an R is presumably a miscut by the mason.Google Scholar

⁷¹ See RSOT, ‘Where the lightning struck: RIB 1426’, in Ant. Journ. lxvi (1986), forthcoming.Google Scholar

⁷² Clayton, J. to Bruce, J. C. (12 Jan. 1863)Google Scholar from Chesters. MS letter in Durham University Library, in Bruce, J. C., Roman Wall ed. 2 (1853) (disbound and interleaved), on verso of first interleaf before p. 141. Information from Mr A. Welfare; inspected by RPW.Google Scholar

⁷³ In LS 159, followed indirectly by CIL vii 623. On the verso of the first interleaf before p. 141 of his Roman Wall ed. 2 (1853) (disbound and interleaved) in Durham University Library, J. C. Bruce noted an inscribed stone ‘found in excavating the bridge’, which he read as [FID]ELIS [VEXIL]LATIO | VIC(TRIX). This inscription, which is otherwise unrecorded, matches Bruce's reading of RIB 1547 LS 159), which is now in Chesters Museum. It is simpler to assume that Bruce mistook the provenance of LS 159, and thus to identify the two inscriptions, than to suppose that the Chesters inscription was lost by Clayton and that Bruce interpreted two separate texts as [FID]ELIS (etc.). RIB 1547 reads [NYMP]HIS, but it can be seen from the line drawing how this could be mistaken for [FID]ELIS; a dedication to the Nymphs would be appropriate at the bridge abutment, cf. RIB 744. AA² v (1861), 143, reports the discovery of ‘an unsatisfactory fragment of an inscription’ during the 1860 excavations, an item otherwise unrecorded which can reasonably be identified with the present text; Bruce's note is accompanied by a letter to him dated 9 June 1860 from J. Fairless which describes a visit to the excavations, RPW and RSOTGoogle Scholar

⁷⁴ Note with sketch-map by J. C. Bruce in his Roman Wall (see previous note), on the interleaf facing p. 212.

⁷⁵ ibid., note written at the foot of p. 229.Google Scholar

⁷⁶ ibid., note on the interleaf facing p. 229. Bruce reads only the first line, as MALEGA, but there seems no reason to doubt the identification.Google Scholar

⁷⁷ ibid., note on the interleaf facing p. 230. In LS 291 Bruce misleadingly simplified the provenance as ‘near Aesica’ (it is nearer Carvoran).Google Scholar

⁷⁸ ibid., note on the interleaf facing p. 243.Google Scholar

⁷⁹ Keppie, L. J. F., ‘Roman Inscriptions from Scotland: some additions and corrections to RIB I’, PSAS cxiii (1983), 391–404. See I.1 with fig. 1 (pp. 391–2).Google Scholar

⁸⁰ ibid., II.2 with fig. 2 (pp. 392–3).Google Scholar

⁸¹ ibid., II.4 with fig. 4 (pp. 393–5). Dr Keppie draws attention to the similarity of decoration between this altar and RIB 1778, an altar dedicated by a prefect of the cohort at Carvoran in a.d. 136–8.Google Scholar

⁸² Offered for sale by Arthur Davidson Ltd., 78/79 Jermyn Street, where it was seen by RSOT. How and when it reached this country is not known. The reading in CIL (‘Vv. divisio incerta’) is slightly inaccurate.