¹ When measurements are quoted the width precedes the height. RIB means Collingwood, R. G. and Wright, R. P., The Roman Inscriptions of Britain, vol. I (Oxford, 1965)Google Scholar, cited by item-numbers. Mr Hassall and Dr Tomlin have undertaken the compilation and editing of this section, and Mr Wright remains available as consultant (see the Editorial in Britannia vii (1976), p. ix).Google Scholar Mr Hassall would like to acknowledge his debt to the unpublished analytical index of personal names in RIB i and CIL xiii prepared by Ian Marriott.

² For the Southwark and Lambeth Archaeological Excavation Committee directed by M. J. Hammerson who, together with Joanna Bird, supplied details of this and the following item. For the sculptures see above, p. 453.

³ L. I will be part of the deceased's name followed by the formula vixit annos probably abbreviated. The age is given in years (36), months and presumably days as in RIB 670, 674, 683, etc. L. 3 may also have included the nomen of the woman who set up the inscription, preceding her cognomen (Matrona), and 1. 5 will have indicated her relationship to the deceased. In 1.5 only the top of a curved letter followed by a serif survives below the first A of Matrona.

⁴ In excavations by the Kent Archaelogical Rescue Unit for the Department of the Environment, directed by B. Philp who made the stone available for inspection.

⁵ Enough survives of the stone in front of the s in 1.2 to show that there was either a gap, or that the preceding letter was probably either A or L.

⁶ During excavations on a building-site (house No. 115), directed by J. Connell who sent details and photographs.

⁷ Or possibly MAT DO. If intended to represent Mars, the inscription might be interpreted as MATRI (for MARTI) or MA(R)T(l) DO.

⁸ By two students (1975-6) of Lancaster University, D. Douglas (in whose possession the stone remains) and I. S. Hargreaves. Dr D. C. A. Shotter sent a photograph and details.

⁹ Only the bottom-third survives of the first letter: if E, then Felix (and cognates) is likely; if L, then Bellicus (etc.). Thus …]LICI could be the end of a personal name in genitive or dative, or even part of a name (whether of deceased or dedicant). PATRI may be preceded by the dedicant's name (RIB 16, 749), or followed by it (cf. 1250), or may be part of a formula like filio patri pientissimo (594).

¹⁰ The stone was noticed by Mr H. G. Ramm. The Rev. P. F. Stirk allowed access.

¹¹ These dedications are not common. For Antoninus Pius the full formula is Divo Antonino (Aug.) Pio, for Caracalla Divo magno Antonino, neither of which seems to have been used here. The formulation may have been similar to that of a dedication to the deified Antoninus Pius at Lambaesis in the reign of Marcus (ILS 9188), Divo Antonino C. Satrius (etc.) dedicante D. Fonteio (etc.), although the lettering and the ligation of Firmin … suggest a later date. Sacrifice is made to both Antoninus Pius and Caracalla as Divo Antonino in the Feriale Duranum ofc. TTAYS (R. O. Fink, Roman Military Records on Papyrus (1971), No. 117, col. ii, 20 and 2 respectively), but in each context the emperor is specified, and elsewhere (col. i, 18) sacrifice is made to Caracalla as Divo Antonino Magno. The formula ob honorem domus divinae occurs nearby at Brough-on-Humber (RIB 707, Antoninus Pius) and elsewhere, but its abbreviated form ob honorem d.d. seems unparalleled, though cf. ILS 4198, in honorem d.d. (Gordian).

¹² During excavation for the Department of the Environment directed by Mr P. S. Austen, who made it and the next item available. They will go to Carlisle Museum.

¹³ For this spelling, see K. H. Jackson, Language and History in Early Britain (1953), 430. Dedications to Belatucadrus (variously spelt, and sometimes identified with Mars) are local to military sites in Cumberland and its immediate borders, where they are common.

¹⁴ During excavation for the Department of the Environment directed by Mr P. S. Austen, who made the stone available. It will go to Carlisle Museum.

¹⁵ See p. 421

¹⁶ The ansate panel and style of lettering suggest an ‘Antonine’ date rather than Hadrianic or Severan, but closer dating seems impossible.

¹⁷ Only one parallel can be found among the 33 building-inscriptions of vexillations of the three British legions collected by Saxer, R., Untersuchungen zu den Vexillationen des römischen Kaiserheeres von Augustus bis Diokletian (Epigraphische Studien 1 (1967)).Google Scholar This too is an inscription of the Sixth Legion, from Corbridge (RIB 1162), and its wording is similar: Legio(nis) Sextae Pi(a)e F(idelis) vex(illatio) refe(cit). But it cannot be dated.

¹⁸ During excavation for the Department of the Environment directed by Mr P. R. Scott, who made this and the next item available. They will go to the Bowes Museum.

¹⁹ This and the previous item were evidently built into the fort wall within c. 3 m of one another, and seem to be from the same hand. Each carries a symbol and the letters P, A, O in two combinations (perhaps AO preceded or followed by p). One symbol is obviously a phallus, which is commonly associated with a personal name (e.g. RIB 631, 872); the other symbol, though suggestive, has not been identified with certainty. The stones are presumably apotropaic or an individual mason's ‘signature’, rather than the record of a building gang's pedatura.

²⁰ By the owner of Peth Cottage, Mr B. Greenwell.

²¹ The left-hand margin is preserved, but it is uncertain whether the stone is otherwise complete. The initial medial stop (?) does not seem to be a centurial sign.

²² During excavation for the City of Newcastle directed by Miss R. B. Harbottle, who provided details of this and the next item. They are now in the Museum of Archaeology, where Dr D. J. Smith made them available.

²³ Line 2: the letters are larger and more widely spaced than in the other two lines, and it is uncertain whether one or two letters are missing both before and after G. The final letter might be D.

²⁴ S (for s(ancto)) may follow Deo, but surviving traces do not fit Cocidio, and the dedication remains obscure. Surviving traces fit Saturninus, but the restoration is conjectural. The final pro suggests the formula pro se et suis, for which however there is insufficient room and little support in the surviving traces.

²⁵ During excavation for Edinburgh City Museums (where the stone now is) directed by Mr N. M. McQ. Holmes, who provided a photograph and details.

²⁶ Only thet op half of 1. 2 is preserved, and the first letter could be G. D M is an abbreviation used of local gods identified with Mars, thus RIB 305 (Nodons) and 1024, D M Condati, Piercebridge. Mars Condates is also attested at Bowes (731) and Chester-le-Street (1045), and seems to be local to north-east Roman Britain.

²⁷ By the excavation section of Grosvenor Museum under the direction of P. J. Davey. We are grateful to P. Carrington of the Grosvenor Museum who provided squeezes and discussed the restoration of the text with us.

²⁸ Owing to shrinkage the letters on (a) which have been overtired are smaller than those on (b). The s in cos has been reversed.

²⁹ A.D. 167. For the rare nomen Viducius, formed from the cognomen Viducus, see Britannia viii (1977), 429, No. 18.Google Scholar

³⁰ During rescue excavation for the Department of vironment directed by Mr P. S. Austen, who made this and the next item available. They will go to Carlisle Museum.

³¹ See p. 421.

³² Possible names include Secundus and Verecundus (and their cognates).

³³ During rescue excaavtion for the Department of the Environment directed by Mr G. Smith (see above, pp. 19-56). Mr P. S. Austen sent the sherd for study. It will go to Carlisle Museum.

³⁴ Presumably the end of a personal name. The M (or ) is preceded by what may be a ligatured v, but is probably only a downstroke.

³⁵ Excavations by the Archaeological Field Unit of the Royal Albert Memorial Museum, directed by P. Bidwell who made this and the following four items available for inspection.

³⁶ The name Hilarus is statistically four times more likely than Clarus.

³⁷ Either Apri, ‘(property) of Aper’, or Apri[lis, ‘(property) of Aprilis’.

³⁸ For the slogan in (a) compare the well-known mosaic pavement at Pompeii (illustrated in J. M. C. Toynbee, Animals in Roman Life and Art (1973), pl. 43. In (b) the first letter is a capital, the second cursive.

³⁹ During excavation for the Bowes Museum and Durham County Council directed by Mr R. F. J. Jones, who made them and the next item available. They are now in the Bowes Museum.

⁴⁰ CIL vii 1234 with Archaeologia xciii (1949), 29.Google Scholar

⁴¹ During excavation for the Bowes Museum and Durham County Council directed by Mr I. M. Ferris and Mr R. F. J. Jones. Mr Jones made it and the next three items available. They are now in the BowesMuseum.

⁴² Hooppell, R. E., Vinovia: a buried Roman city (1891), 40Google Scholar, with Haverfield's improved reading in Arch. Journ. 1 (1893), 293Google Scholar: Armea me docuit [….

⁴³ The reference is obscure, but the formula me docuit is a conventional way of saying one's attitudes have been changed by falling in love with some woman: cf. Propertius I, 10, 19, Cynthia me docuit semper quaecumque petenda ∣ quaeque cavenda forent, and CIL iv, 1520, Candida me docuit nigras odisse puellas. The personal name Armea is otherwise unknown.

⁴⁴ For names ending with the common Greek suffix-ανδροѕ, see Bechtel, Historischen Personennamen des Griechischen (1917), 49-52.

⁴⁵ Excavations by the Colchester Archaeological Unit for the Colchester Excavation Committee, directed by P. Crummy. Nina Crummy supplied full information and made this and the following ten items available for inspection.

⁴⁶ By W. J. and Kirsty Rodwell who sent rubbings and provided details.

⁴⁷ For Mars Toutatis cf. the inscribed silver plate from Barkway (Herts), RIB 219, and the note loc. cit.

⁴⁸ Information and drawings from the excavator, B. Rawes.

⁴⁹ See Clifford, Elsie, JRS xlv (1955), 68–72.Google Scholar

⁵⁰ By H. J. M. Green who supplied details and a rubbing.

⁵¹ A variant of the classical Latin cupa, ‘tub’, or ‘vat’. In medieval Latin the meaning is ‘cup’, cf. J. F. Niermeyer, Mediae Latinitatis Lexicon Minus (1976), s.v. cuppa.

⁵² During excavations by the Kent Archaeological Rescue Unit directed by B. Philp who made the tile available for inspection, and supplied details and a drawing.

⁵³ Excavations directed by M. Biddle. Information from Fiona Gale who submitted the sherd for inspection.

⁵⁴ During excavations for Leicestershire Museums and the Department of the Environment by the Leicestershire Museums Archaeological Field Unit directed by Jean Mellor. Full information, drawings and photographs for this and the following item from P. Clay. For other sealings from Leicester, see Britannia vii (1976), 387Google Scholar, Nos. 36-40, with corrigenda ibid, viii (1977) 449.

⁵⁵ The Ala Vocontiorum is attested in the British diploma of 122 (CIL xvi, 69), and at Newstead by RIB 2121. For another seal possibly relating to this unit from South Shields see Richmond, I. A. in AA ⁴ xi (1934), 102Google Scholar, No. 23: OVA (A(la) Vo(contiorum) retrograde?).

⁵⁶ Excavations for the Museum of London by the Department of Urban Archaeology directed by D. Jones. M. Rhodes of the Department provided full details and Xerox copies of drawings of this and the following nine items.

⁵⁷ Curie, J., Newstead, A Roman Frontier Post and its People, 314Google Scholar, pl. XXXIII, No. 2. This parallel was pointed out to us by H. Chapman of the Museum of London.

⁵⁸ Inscribed IHIS CONICILLI IKI. See Woodward, A. M., ‘The Roman Fort at Ilkley’, Yorks. Arch. Journ. xxviii (1926), 289 and fig. 49.Google Scholar

⁵⁹ Germania Romana, 2nd ed. pt. 5 (1930), 13 and pl. xvi, 5.

⁶⁰ Kajanto, I., The Latin Cognomina, cites only one example, CIL ii, 5964Google Scholar, Tarraconensis. Note also a second example occurring as a graffito on a cup, Drag. 27 found in Roman London, JRS xliv (1954), 109Google Scholar, No. 38, where the name has been unnecessarily expanded to Eques(ter).

⁶¹ Read independently by M. Rhodes of the Department of Urban Archaeology as [.]MIṂ.

⁶² The longer inscription (a) cannot be matched in the unpublished index of dipinti on amphorae from Rome and Pompeii prepared by I. Marriott. The same index shows that the phrase Gari Flos, ‘first-quality fish sauce’ is abbreviated to G.FLOS and G.F., and on this analogy G alone could stand for Garum. In that case (b) could conceivably stand for G(ari) IIII C(ongii), ‘4 congii of garum’, that is 13.13 litres. This however, as C. Green of the Department of Urban Archaeology informs us, is less than would be expected of a full vessel of this type (20-30 litres), and the reading of (b) is anyway not certain.

⁶³ The numeral 7 with or without a fraction occurs fairly commonly as an indication of capacity on amphorae found in Britain, but the unit of measurement is uncertain. We are informed by C. Green, however, that amphorae of Dressel form 20 hold between 60 and 80 litres.

⁶⁴ Excavations by the Central Excavation Unit of the Department of the Environment directed by J. Hinchliffe. Information on this and the following two items from N. D. Balaam who made the objects available for inspection.

⁶⁵ Kajanto, I., The Latin Cognomina, cites 2,430 examples of Fortunatus in CIL (836 from Africa), as opposed to 130 of Fortis.Google Scholar

⁶⁶ If this is the end of a personal name in the dative, Avitius, Equitius and Tarquitius are possibilities.

⁶⁷ By Fit. Lt. R. A. Brown who provided details and a Xerox of a drawing. For the general type see Boesterd, M. H. P. den, Description of the Collection in the Rijksmuseum G. M. Kam at Nijmegen (Nijmegen, 1956), 7–11Google Scholar, Cat. Nos. I5a-29a, pls. I and II.

⁶⁸ Information and photograph from A. E. Rollings. The ring remains in the possession of the finder. For the site see Bulletin of the Northants Federation of Arch. Socs. i (1967), 7Google Scholar; ii (1967), 7; iii (1969), 6; iv (1970), 6.

⁶⁹ For an example of the same exhortation on a ring see EE ix 1334 (Kilbride, Lanarks.).

⁷⁰ During excavation for the Vindolanda Trust directed by Mr A. R. Birley, who made this and the next two items available.

⁷¹ For such sealings, see Richmond, I. A., Cumb. Westm. n.s. xxxvi (1936), 104 ff.Google Scholar This one is not the abbreviated title of a cohort, since there is no stroke above the v, and no Fifth Cohort of A … is known in Roman Britain. Compare CVP (Richmond, p. 118), and another sealing from Brough-under-Stainmore now in the British Museum (accession No. 74, 12-28, 85),which has been stamped on the obverse bv a rectangular die, c[.] A.

⁷² The stamp, presumably the maker's name, is divided into two panels, perhaps by damage to the die. The first half (c. 4 letters) is illegible. We have not been able to find a comparable stamp.

⁷³ For stamped bricks and tiles of the Sixth Legion already found in the bath-house, see Britannia vii (1976), 390.Google Scholar

⁷⁴ As a surface find ‘within 30 miles of Brandon’, subsequently acquired with other finds from Norfolk and the Norfolk/Suffolk borders by Mr J. L. Homan, who made it available. R.S.O.T.

⁷⁵ For the interchange of η and ϵ in Koine Greek, see Gignac, F. T., A Grammar of the Greek Papyri of the Roman and Byzantine Periods, i (Milan 1976), 242 f.Google Scholar, esp. 245, ϵ>η as here, before a liquid or in final position. συνέρΝομαι (‘come together’), of which σύνϵλϴϵ is the strong aorist imperative, is common in Koine Greek in the sense of ‘marry’: Preisigke, F., Wörterbuch der griechischen Papyrusurkunden ii (Berlin, 1927)Google Scholar, s.v.

⁷⁶ Excavations for the Sussex Archaeological Society by J. H. Money who supplied details and rubbings.

⁷⁷ Found by the farmer Mr S. Spence in field No. 2718, north of Blind Lane and reported by him to K. Scott. Information, photograph and Xerox of drawing (by Diana Webster), supplied by G. Webster.

⁷⁸ If alphabetic this could be a palindromic charm, see K. Preisendanz, Papyri Graecae Magicae vol. 3, index XIII, palindromes, and A. Audollent, Defixionum Tabellae, index VII, for numerous examples; C. Bonner, Studies in Magical Amulets, 202-4, ‘Palindromes’, holds that it is wrong to try and wring sense from most of them.

⁷⁹ During excavation for the University of Lancaster and the Department of the Environment, directed by Dr T. W. Potter, who sent the piece for study. It will go to the Borough Museum, Kendal. For the site, see Cumb. and Westm. ² lxxvi (1976), 6 ff.Google Scholar

⁸⁰ The stamp appeared only after chemical stripping, and its letters are badly corroded; but what survives, with the shape of the handle and the use of a decorative motif of pairs of concentric circles arranged in a triangle, accord with a handle-end bearing this stamp from the fort of Stockstadt (ORL B iii (1914), No. 33, p. 53 and pl. vii, 54Google Scholar, cf. CIL xiii, 10027, 13a (Châlonsur-Sâone)).

⁸¹ During excavation for the Richmondshire Excavation Group-directed by Mrs S. Thubron, who made them available.

⁸² For manu in this context, see Britannia ii (1971), 303, No. 91Google Scholar; and JRS lii (1962), 199Google Scholar, No. 53, ϕε (i.e. 505 items), manu Charmidanax. For notes of weight, cf. JRS xxix (1939), 229Google Scholar, No. 15, and CIL xiii, 10008, 43 f. Elaborate tallies survive in considerable numbers from the kilns at La Graufesenque (Marichal, R., ‘Nouveaux Graffites de la Graufesenque’, Revue des Etudes Anciennes 76 (1974), 85 f.)CrossRefGoogle Scholar, but are otherwise uncommon. For a tilemaker's itemised tally, see JRS xxx (1940), 187–8, No. 20.Google Scholar

⁸³ During excavation by the Scottish Field School of Archaeology directed by Professor S. S. Frere. The sherd was made available by Miss M. G. Wilson.

⁸⁴ During excavation for the Department of the Environment directed by Mr P. J. Casey, who made it avanaoie, and by Mr J. L. Davies. It will go to the National Museum of Wales, Cardiff. For the discovery see Britannia viii (1977), 358 and pl. XXII B.Google Scholar

⁸⁵ Casual scratches, and the poor quality of the letters, prevent a certain reading. It might be , but there are other possibilities. Neither practice lettering nor a personal name seem likely explanations.

⁸⁶ Excavations for the Clwyd-Powis Archaeological Trust directed by T. O'Leary. Information from P. Webster who submitted these tiles and the following items for inspection, and sent a rubbing of (c). For illustrations of the tile-stamps found at Holt see W. F. Grimes, Holt: The works depot of the Twentieth Legion at Castle Lyons, 139-43 and figs. 59 and 60 (No. 3). For the site see above, p. 406.

⁸⁷ In 1. 2 the last letter could be an A and in 1. 4 the letter could be x. The graffito could represent the tria nomina of two individuals, as Quintus Avid(ius?) Amabilis, Ạ[u]∣Ịus Antḷ[stius] ∣ Feli]x. The rare praenomen Vibius, though theoretically possible as an alternative to Aulus, is much less likely.

⁸⁸ Excavations for the Department of the Environment and University College, Cardiff directed by W. H. Manning. Information on this and the following fourteen items from Sabina Thompson who made the sherds available for inspection.

⁸⁹ Or possibly ‘(property) of Rufius’, but this nomen is rare (cf. CIL vi, 2587 (from Volaterrae) and xi 2714 (from Volsinii)).

⁹⁰ In (b) the x has been cut first, and was perhaps originally intended as a numeral. The letter M has then been added within the angle formed by two of the arms of the x, the cross stroke of the A has finally been cut in the adjacent angle with two arms of the x' acting as its sides.

⁹¹ If V is alphabetic rather than numerical this could be an abbreviation for a name such as SV(PER) or SV(PERSTIS).

⁹² The letters are cut tangentially to the footring.

⁹³ The Greek personal name Nicephoros is not uncommonly spelt Niceforus (see CIL vi, Part vii, fasc. 4). As an adjective it is also the regular translation of the Latin word victrix, one of the titles of the Twentieth Legion (cf. ILS 2657).

⁹⁴ The occupant, Mr John Featherstone, informed Georgina Plowright, Assistant Curator in Archaeology and History at the City of Bristol Museum and Art Gallery that the stone was included in the landlord's inventory when he moved in. We are grateful to Mr Featherstone for a photograph and to Miss Plowright who reported the piece (to R.P.w.).

⁹⁵ By Mr S. Rammell who made it available through Mr L. P. Wenham. Professor D. Fishwick commented on the questions it raised concerning the imperial cult.

⁹⁶ (I) The ES axe of un-Roman form, and the omission of the final s of Caes(ar) and the division of deus are epigraphically implausible. (2) First-century emperors when deified were styled divus not deus, and never described as the god of a province. (3) It is unlikely that Claudius would have been described as a god in his lifetime in a western province, or that the temple at Colchester was completed before his death (see Fishwick, , Britannia iii (1972), 164 ff.).CrossRefGoogle Scholar (4) The (un-Latin) phrase deus Britanniae echoes not only Apocolocyntosis 8, but also (especially as divided de – us) the stamp on the earliest surviving Roman lead ingot from Britain: Ti. Claudius Caesar Aug. (etc.) de Briton (CIL vii, 1201); and perhaps also the inscription De Britann on Claudius's triumphal arch as depicted on a well-known aureus in the British Museum (BMC, Imp. I, p. 168, No. 29).

⁹⁷ ‘[Claudius] wants to become a god. Is it not enough that he has a temple in Britain, that the barbarians worship him and pray to him as a god?’ Written immediately after Claudius's death, perhaps by Seneca. The forger, oddly enough, does not seem to have been influenced by Tacitus, , Ann. xiv 31Google Scholar, ‘templum divo Claudio constitutum’. R.S.O.T.

⁹⁸ By R.S.O.T., following suggestions made independently by Miss J. M. Reynolds and Mr R. Lane Fox. We are grateful to Miss Reynolds for discussing the reading with us, and for consulting Mr R. G. G. Coleman and Dr W. H. Plommer, whose opinion we have followed. Professor J. E. Bogaers will be discussing the inscription, with a restoration differing in some details from that proposed in Britannia viii, in Oudheidkundige Mededelingen uit het Rüksmuseum van Oudheden te Leiden lviii (1977)Google Scholar, forthcoming.

⁹⁹ The spacing and surviving traces of the initial letter of IANVM require 1 not F (the subscript dot was omitted in printing). The first stroke of the M, which is not properly visible in ibid., pl. XXVIII B, has now been cleared of the mortar which filled it. It does not allow room for an A after the v (thus ianuam, ‘a door’), unless it was ligatured with a cramped and near-vertical second stroke; of this there is no trace, and its absence is implied by the placing of v almost midway between AN and M. For ianus in the sense of ‘archway, gateway, covered passage’, see OLD s.v., and Daremberg-Saglio, Dictionnaire s.v. Ianus II. A ianus might be an urban landmark spanning a road (e.g. ILS 5384, 8345); there seems to be no epigraphic evidence of its association with an arcuś, but cf. Suetonius, Domitian 13, ‘ianos arcusque cum quadrigis et insignibus triurnphorum per regiones urbis … exstruxit’, where two different but related structures seem to be meant. Viducius apparently paid for an arch giving on to a covered passage-way.

¹⁰⁰ For parallels she cites Berger, L., Römische Gläser aus Vindonissa (Basel, 1960), 56–67, pls. ix and x.Google Scholar