² Information from P. Marsden who directed excavations for the Guildhall Museum.

³ Information and drawing from G. Dawson who directed excavations by the Southwark and Lambeth Archaeological Society for the Southwark Archaeological Excavation Committee. The site lies in the angle formed by Tooley Street and Borough High Street.

⁴ During excavations for the Canterbury Archaeological Trust directed by T. W. T. TattonBrown who submitted the object for inspection. For the site, see p. 423 above.

⁵ Preceding the s is a line which could be the attenuated and slightly curving horizontal stroke of an L. Below the s, and on the very edge of the sheet, is a mark which could possibly be the top of the letter s or c.

⁶ Rescue excavations for the Kent Archaeological Rescue Unit and for the Department of the Environment were directed by B. Philp, who made the altar available for inspection. See Kent Archaeological Review 47 (1977), 170–1 with plateGoogle Scholar.

⁷ In 1. I only a small segment of the curve of the o survives. In 1. 3, after the stop, is the serif from the top of a letter which rules out curved letter forms and, on spacing, the letter A.

⁸ The text has been phrased in an unusual order; for examples where the deity is put in a secondary position see RIB 140 (Bath) and for the Matres 652 (York). In 1. 4 it is rare to find the letter 1 used both in the ligatured RIB and also as the initial letter of ITALICIS. Note also: L. I st(rator) co(n)s(ularis): see RIB 233 from Irchester, Northants. These officers had charge of the establishment of horses and the like, here for the provincial governor. Dover, as the starting point of a trunk route, would be an appropriate place, R.P.W. L. 2 OL(VS) is the rare form of the praenomen Aulus. See ILS 1932 (Republican)CIL v, 6445 (Ticinum) and ILS 3923 (CIL iii, 993), (Apulum, probably Severan). As nomen CORDIVS, with small 1 inside D, would fit, see ILS 902,2452, 5947,1. 26, and TLL Onomasticon (several examples), R.P.W. LI. 3–4 De Ruggiero, Dizionario s.v. Lares, cites no title Italici. But for Matres see RIB 88 (ILS 4786), Winchester, Matres Italae, etc., and RIB 653 (ILS 4787), York.

⁹ During excavations for the Cirencester Excavation Committee directed by A. D. McWhirr. Interim report in Antiq. Journ. forthcoming. The inscription was made available for inspection by Valery Rigby.

¹⁰ By E. J. Swain. Information initially from P. Arthur. For other epigraphic finds from the site see Nos. 79–82 below.

¹¹ By Mr H. Morris who has presented it to the Colchester and Essex Museum, information and photograph from G. M. R. Davies, assistant curator at the Museum.

¹² For the temple see M. R. Hull, Roman Colchester, 224–33, and M. J. Lewis, Temples in Roman Britain, under ‘Colchester 2’.

¹³ For three very similar plaques from Colchester see RIB 191, 194 and 195, and for the nomen Oranius CIL ix 4686 (Reate) and CIL x 4262 (Capua).

¹⁴ Excavations for the Colchester Archaeological Trust were directed by P. Crummy, who supplied full information.

¹⁵ This formula, though not common in Britain, does occur, e.g. RIB 16, 87, 672 and 677.

¹⁶ In 1. 2 there will have been room for the filiation following the deceased's nomen. In 1. 3 although RO[M(ILIA)], the voting tribe, is theoretically possible, a cognomen such as RO[MANUS] or [RO]GATUS is more likely. In 1. 4 there is room for a single wide digit such as x or v at the beginning of the line, or two narrow ones. Since there is a bar above this numeral it is unlikely to be the deceased's age or years of service, though superscript bars are occasionally used with both (RIB 265 and 482, RIB 257). If it is the number of a unit, the type of unit and deceased's rank will have preceded it at the end of 1. 3. To avoid crowding, it will probably be necessary to assume that he held the rank of centurion, indicated by a centurial sign.

¹⁷ Excavations for Lincoln Archaeological Trust were directed by Christina Colyer. Information from Miss Colyer and the assistant director M. J. Jones who excavated the site, on which see p. 390 above.

¹⁸ It is just possible that an original text has been erased, or if this is the original text that the letters have been recut. Note especially the faint diagonal cut preceding the N in 1. I, but too close to it to be the left side of the letter A, and the faint vertical following the R in 1. 2.

¹⁹ For the proposed restoration in 1. 1 compare CIL vi 7386, 14233 and 22098, examples of the common formula Dis Manibus Sacrum abbreviated in this way. The deceased's praenomen and nomen will have occurred in 1. 2, and his cognomen in 1. 3, while the origo will fall naturally in 1. 4, where Joyce Reynolds proposes do]mo Sa[., the town name being perhaps abbreviated as, for example, Sav(aria), RIB 258 (Lincoln). In 1. 6 the word decurio suggests itself, but if this was the rank held by the deceased, it would be more usual for the office to precede his age, which is, apparently, given in 11. 5 and 6.

²⁰ The fragment is buff on the upper side, but reddened on the lower side. This part of the cemetery is only now coming into use. Two late secondto early third-century cremation-urns were found close by; some inhumations have been disturbed in digging other graves. The fragment is now in Lincoln City and County Museum; Mr A. J. White sent details and a tracing. For other finds see Richmond, , Arch. Journ. ciii (1947), 52Google Scholar. R.P.W.

²¹ By Mr and Mrs D. Wilson of Bryn Eglwys, Llanddyfnan, Anglesey, in whose possession the stone remains. Details were sent by Mr C. Hughes, who is publishing a note in the Lincolnshire Historical and Archaeological Journal, forthcoming. R.S.O.T.

²² By A. W. J. Houghton who provided full information. The stone belongs to the National Trust on whose property it was discovered, and is at present at the Site Museum at Wroxeter.

²³ Recovered by J. K. Knight of the Ancient Monuments Inspectorate of the Department of the Environment. Information and squeeze from G. C. Boon, Keeper of Archaeology in the National Museum of Wales, Cardiff, where the stone now is.

²⁴ The sites lies in the vicinity of one of the legionary cemeteries (see RIB 359). Rescue excavation was conducted by B. Vyner of the Glamorgan and Gwent Archaeological Trust and G. C. Evans of the National Museum of Wales. The inscription and the following item were presented to the Museum by the land-owner, P. F. L. Phillips. For the site, see p. 360

²⁵ In 1. 2 the first letter could be R or B, and the last a G or an o.

²⁶ Although cognomen Oneratus is attested, it is very rare compared to the extremely common (H)onoratus/a (two examples of the former, 666 of the latter listed by Kajanto in The Latin Cognomina), and it is possible that Honorata is intended. Mann, J. C., ‘Spoken Latin in Britain as evidenced in the Inscriptions’, Britannia ii (1971), 218–24CrossRefGoogle Scholar, however, gives no examples of E used instead of o, although he cites five examples, all in proper names, where o is used instead of E.

²⁷ During excavations directed by D. Mason for the Excavation section of the Grosvenor Museum who supplied a photo and Xerox of drawing. For the site, see p. 387.

²⁸ In 1. 5, an original letter N seems to have been altered to an M; however, Commitus, if this is the correct reading of the name of the dedicator, appears to be unmatched.

²⁸ Excavation by J. Hinchcliffe of the Central Excavation Unit of the Department of the Environment. For the site see p. 385, and for another epigraphic item No. 49. Information from N. Balaam of the Central Excavation Unit.

²⁹ During rescue excavation directed for the York Archaeological Trust by Mrs S. Donaghey. Mr A. MacGregor made the stone available; it will probably go to the Yorkshire Museum. We would like to thank the scholars who have discussed this text with us, particularly Prof. J. E. Bogaers, Mr B. R. Hartley, and Dr J. C. Mann.

³⁰ Stuart, P. and Bogaers, J. E., Deae Nehalenniae (Middelberg/Leiden 1971)Google Scholar, No. 45: Deae Nehalenniae Placidus Viduci fil(ius) | cives Veliocassinius | negotiat(or) Britann(icianus) | v(otum) s(olvit) l(ibens) m(erito).

³¹ Three letters are guaranteed by Placidus (1. 3), and a minimum of five by et numinibus Augustorum (1. 2). However, praenomen and nomen are unlikely to be separated from the cognomen by more than a patronymic (a tribus being unlikely in this late inscription); its presence is implied by the reference to Viducius's father on the Nehalennia dedication. L(ucius) is the most likely praenomen.

³² Reading: I▲O▲D▲ (1. 1), ET▲NN▲AV (1. 2), L▲F▲PLA (1. 3), CIVIT▲ (1. 4), VIVIR▲N (1. 5), CRET▲(1. 6), D▲D▲L▲D▲D▲D▲ (17).

³³ Notes on the restoration. L. 1: Dedications genio loci (RIB 646, 647) and numini Augusti et genio Eboraci (RIB 657) have been found at York, but no deity is regularly coupled with the imperial numina and genius loci. There is a space of about five letters. The cult of Jupiter Dolichenus in Britain was almost confined to the northern frontier, but it may have extended to York because of the legion (cf. RIB 1131, a dedication by one of its centurions at Corbridge): there is a contemporary dedication at Piercebridge (RIB 1022, A.D. 217). Negotiatores contemporary with Viducius made dedications to Jupiter Dolichenus (ILS 9286 Mainz, A.D. 217, cf. 4322, Rome, A.D. 218). L. 2: The emperors in 221 were Elagabalus and Severus Alexander (still only Caesar), but a dedication of this year at Rome (CIL vi 2999) refers to them jointly as Augusti, cf. RIB 1466. LI. 3–4: this use of the domo formula is difficult to parallel since it is regularly used either with a tribal adjective in the nominative or with the name of the town (in the ablative, or sometimes genitive/locative) unqualified by the name of the tribe. Domo seems to be unattested with civitas + tribal name. However, civitas Veliocassium should be preferred to Rotomagus Veliocassium, since this form does not seem to have been used (and Rolom(agi) would not fit the space available).Civi7as Veliocassium is the usual form (see CIL xiii, 1, p. 512), and is implied by Viducius's description of himself as cives Veliocassinius (cf. CIL xiii, 1717, cives Velioca[…]. Another negotiator from Rouen states his origin as ex civitate Veliocassium (ILS 7035, cf. 9476). L. 5: Viducius is likely to have been a sevir, since he was a successful merchant with connections outside York, like the two known seviri Lunaris, M. Aurelius (JRS xi (1921), 102)Google Scholar and M. Verecundius Diogenes (RIB 678 with JRS lvi (1966), 228)Google Scholar, both with Aquitanian connections.

L. 6: Viducius describes himself as negotiator Britannicianus, but is unlikely to have used this adjective within Britain itself. Gallic negotiatores often specify the commodities in which they dealt (CIL xiii, 5, pp. 160–1), and there were dealers in salt fish (salarius), fish-sauce (allecarius), and wine (vinarius) at the shrine of Nehalennia. Another shrine of Nehalennia was patronised by a negotiator cretarius Britannicianus (ILS 4751 = Stuart and Bogaers No. 6), a particularly attractive parallel, in view of the high proportion of East Gaulish samian found at York (see R. M. Butler (ed.), Soldier and Civilian in Roman Yorkshire (1971), 131). L. 7: a formula is required whose penultimate letter is D, and D.S.P.D.D. (etc.) is too short. The colonia of York had decurions (RIB 674).

³⁴ JRS xi (1921), 102Google Scholar, cf. RIB 678, both cited in the previous note. For the trading connections of Roman York, see R. M. Butler, op. cit. (note 33), 127 ff.

³⁵ The stone was known to the grandmother of the present owner, Mr R. Hunter, who informed Mr P. S. Austen, and allowed access. Obscuring mortar was removed. The centurion is otherwise unknown, and the restoration, probably of a cognomen (Pontianus, etc.) is uncertain.

³⁶ During excavation for the University of Newcastle and the Department of the Environment directed by Mr C. M. Daniels, who made it available, and by Mr J. P. Gillam. It is now in the Museum of Antiquities, Newcastle. For the site, see p. 372.

³⁷ The M has been partly erased. The centurion is otherwise unknown, but his nomen is attested in Italy.

³⁸ During the same excavation as No. 20. Mr Daniels made it available. It is now in the Museum of Antiquities, Newcastle.

³⁹ The c is preceded by a triangle of three short vertical ‘pecks’, which cannot be interpreted as a centurial sign. For building-stones with personal names in this area of the Wall, see RIB 1640 and JRS lii (1962), 194Google Scholar, No. 18. The name Cunaris seems to be unrecorded, but is acceptable as a Celtic formation.

⁴⁰ During excavation directed for the Vindolanda Trust by Mr R. E. Birley, who made it and Nos. 23-26 available. He has published them in Vindolanda I. 1976 Excavations, Interim Report (1977), 37. They are now in the site museum.

⁴¹ This is the eighth ‘Veteres’ altar from Chesterholm. For correctness of the spelling with H, see note to RIB 1603.

⁴² Reconstruction in 205/7 of forts associated with Hadrian's Wall is already attested from Corbridge (RIB 1151), Risingham (RIB 1234), Chesters (RIB 1462), Birdoswald (RIB 1909), and perhaps at High Rochester (RIB 1277) and Housesteads (RIB 1612).

⁴³ It does not accord with any inscribed fragment yet found at Chesterholm.

⁴⁴ During renovation of the building (now known as Glenwhelt Coachhouse) by the owner, Mr R. Archer, who told Mr R. E. Birley and allowed access. The stone remains in his possession, built into a fireplace.

⁴⁵ The centurion is already attested in the sector 37–42A (RIB 1652) and near Turrets 48A and 48B (RIB i860, 1862).

⁴⁶ The stone was noticed by Mr P. S. Austen when inspecting RIB 1844, which is built into the adjoining farmhouse (now owned by the Rev. T. Blakey). The byre remains in the possession of Mr L. Scott, Thirlwall Castle Farm, who allowed access. The centurion is otherwise unknown, and so apparently is his cognomen, which is formed from Janus. See I. Kajanto, The Latin Cognomina (1965), 211 ff., for cognomina formed from gods' names, esp. Neptunalis, and Ianilla from Janus (CIL iii 9031).)

⁴⁷ During excavation directed for the Department of the Environment by Mr G. Smith. The stone was made available for study by Mr P. S. Austen, and is now in Carlisle Museum. For the site, see p. 373.

⁴⁸ A photograph and full details were sent by Mr G. S. Maxwell. The stone is now in the National Museum of Antiquities, Edinburgh.

⁴⁹ Sabinianus is otherwise unknown, except for an altar (RIB 2132) found nearby, and now lost, dedicated to the Celtic god of healing, Apollo Grannus.

⁵⁰ The first letter might be E, but a personal name ending in Jeara is unlikely. For nuntium as ‘a (divine) message’, see Oxford Latin Dictionary s.v. Ex nuntio would be an apparently unrecorded variant of the common formula ex visit, ex iussu. etc.

⁵¹ During excavation directed for the Cumbernauld and Kilsyth District Council by Mr H. Millar. The stone is now in Colzium House, Kilsyth. A photograph and full details were sent by Mr L. J. F. Keppie.

⁵² In the margin of his own copy of Britannia (1607) against the entry for Cadder (and thus located in RIB), but without explicit attribution, and only just above the entry for Miniabruch [Kilsyth]. The coincidence of the surviving letters and line-division, and probably of location, makes the identification certain. It was made by Mr Keppie, who has published a note in the Glasgow Archaeological Society Bulletin, spring 1977.

⁵³ Mr Keppie notes that Camden knew of a stone from Bar Hill (RIB 2172) which was taken to ‘a nobleman's house’ at Kilsyth; and that an altar dedicated to Silvanus (RIB 2167) has been found at Bar Hill. Its find-spot is only 3 km south of Kilsyth Castle, whereas Cadder is 12 km to the south-east.

⁵⁴ During excavation for the Department of the Environment directed by Dr D. J. Breeze, who provided a photograph and full details. The stone is at present in Argyle House, Edinburgh. For the site see p. 365.

⁵⁵ There seems to have been no centurial sign. The centurion is otherwise unknown; and since many cognomina begin with Quint…, no expansion is possible.

⁵⁶ Excavation for the Ampthill and District Archaeological and Local History Society was directed by K. J. Fadden, who made this and the following item available for inspection.

⁵⁷ During excavations directed by J. McPeake of the Excavation Section of the Grosvenor Museum. We are grateful to Margaret Buhner for discussing the text with us, and to the staff of the Grosvenor Museum, in particular P. Carrington, for making this and the following twelve items available for inspection. For the site, see p. 385.

⁵⁸ For the nomen Vanius see CIL xi 4314 (and for Vannius CIL xiv 283), and for the cognomen Setinus/a CIL viii 4748 and 22972, although Se(x)tinus may be intended. For iumenta dossuaria see Varro, De Re Rustica ii, 10, 5. Previous finds of inscribed tags from Chester come from Grey Friars (EE vii, 289, No. 878) and the Deanery field (JRS xxi (1931), 250, No. 12Google Scholar).

⁵⁹ During excavations directed by J. McPeake for the Excavation Section of the Grosvenor Museum. We are grateful to Glenys Lloyd-Morgan of the Museum for discussing the paterahandle with us. For the type see Boesterd, M. H. P. Den, Description of the Collections in the Rijksmuseum G.M.Kam at Nijmegen (Nijmegen 1956), 7–8, Catalogue No. 15(a) and pl. IGoogle Scholar.

⁶⁰ See note 57 (ink.). For a similar inscribed disc from the Deanery field, Chester, see JRS xiv (1924), 246, No. 14Google Scholar.

⁶¹ See note 59 (init.).

⁶² For the site and finds see Chester Archaeological Bulletin v (1977), forthcomingGoogle Scholar.

⁶³ See note 57 (init.).

⁶⁴ For the site and finds see Chester Archaeological Bulletin v (1977), forthcomingGoogle Scholar.

⁶⁵ See note 57 (init.).

⁶⁶ Margaret Bulmer points out that the graffito has been cut on a place where the surface of the vessel has been rubbed away as if to erase another name.

⁶⁷ Excavations by T. J. Strickland, Field Officer, the Excavation Section of the Grosvenor Museum.

⁶⁸ See note 59 (init.).

⁶⁹ There is an oblique stroke through the centre of the s. If this is a measure of capacity the M cannot stand for modii, for this would give too great a volume (1 modius = 8754 litres or 16 sextarii, 1 sextarius = 0·547 litres or just under 1 plnt), but possibly mensura, ‘measure’ or ‘capacity’. The unit of measurement would be the congius = 3-283 litres or 6 sextarii. Compare capacities recorded on storage-vessels or amphorae found at Birrens C(pngii) in q(uartarii) v Kjgulae) ii = 18½ plnts, or 10·58 litres (JRS xlii (1952), 108Google Scholar, No. 33) and at Tower 16b on the Cumbrian coast, C(ongii) iii ky(athi) ii = nearly 17½ plnts or 9·94 litres (JRS xlv (1955), 148Google Scholar, No. 25). In No. 87 below M is similarly interpreted mensuravit.

⁷⁰ See note 62.

⁷¹ CIL iv 6117 and C1L xv 3889. We are indebted to the word-index of ‘diplnti’ prepared by I. Marriott for these references.

⁷² Excavations directed by D. Mason for the Excavation Section of the Grosvenor Museum.

⁷³ See note 57 (init.).

⁷⁴ For the type see J. P. Gillam, Types of Roman Coarse Pottery Vessels in Northern Britain, No. 269.

⁷⁵ See note 57 (init.).

⁷⁶ Excavation by J. Hinchcliffe of the Central Excavation Unit of the Department of the Environment. For the site see p. 385 and for another eplgraphic item, No. 17 above. Information on this and the following items from N. Balaam and Gillian Walker of the Central Excavation Unit.

⁷⁷ The first two letters are definitely a repeated v and not x as in (leg) XX VV.

⁷⁸ Information from P. J. Drury who directed the excavations for the Chelmsford Excavation Committee.

⁷⁹ Excavations b y the Colchester Archaeological Unit directed for the Colchester Excavation Committee by P. Crummy who supplied a photograph of a cast. Mr Crummy also supplied details of the following sixteen items, all discovered on sites excavated by the unit under his direction.

⁸⁰ CIL vii 1273, and J. M. C. Toynbee, Art in Britain under the Romans (1964), 377, pl. LXXXVII. Note, however, that here there is a secondary frieze between the inscription and the main chariot frieze.

⁸¹ Visual identification only.

⁸² Visual identification only.

⁸³ For the form and fabric of the vessel and the style of lettering of the graffito compare No. 65 from the Lion Walk site. Both are probably products of the same kiln.

⁸⁴ Either a Greek name such as Nicander or Nicanor or a Celtic name for which several possibilities are cited by Holder, Alt-celtischer Sprachschatz.

⁸⁵ By R. Scott after the completion of rescue-excavations by Rosalind Niblett. Mr Scott presented the find to the Colchester and Essex Museum.

⁸⁶ For the form and fabric of the vessel, and the style of lettering of the graffito, compare No. 61 from the Balkerne Lane site. Both are probably products of the same kiln.

⁸⁷ During excavations by the Atomic Weapons Research Establishment (Foulness) Archaeological Society directed by D. J. and M. R. James. Information, photo and Xerox of drawing supplied by E. F. Hyde.

⁸⁸ Information from W. J. and Kirsty Rodwell who excavated for the Department of the Environment and Essex Archaeological Society. For the site see Britannia iv (1973), 115–27Google Scholar.

⁸⁹ According to I. Kajanto, The Latin Cognomina, 174, Quintasius occurs only on African inscriptions (fifteen examples including two Christian). The name Quintaes, however, presumably a cognate form, is attested at Rome (C1L vi 9245, an example not cited by Kajanto) of which Quint(a)esius could be a variant.

⁹⁰ Rescue excavations by Fowler, P. J., see BGAST xc (1971), 50Google Scholar, fig. 10 and p. 52; and, for the site, Britannia i (1970), 295Google Scholar. The tiles are in Gloucester City Museum where M. J. Watkins made them available for inspection.

⁹¹ During excavations for the Cirencester Excavation Committee directed by A. D. McWhirr, interim report in Antiq. Journ. forthcoming. This, and the following six items were made available for inspection by Valery Rigby.

⁹² See note 91.

⁹³ During excavations for the Cirencester Excavation Committee directed by J. S. Wacher. For the site see Antiq. Journ. xlii (1962), 3–14Google Scholar.

⁹⁴ If this refers to the intended contents there may be some connection with cici, the Egyptian castor-oil tree from which oleum cicinum, castor-oil was produced. Celsus, De Medicina v, 24, 3, Pliny, , Natural Histories xxiii, 41 (section 83)Google Scholar.

⁹⁵ During excavations for the Cirencester Excavation Committee directed by A. D. McWhirr. Interim report in Antiq. Journ., forthcoming.

⁹⁶ See note 91.

⁹⁷ See note 93.

⁹⁸ The last three letters might stand for n(umero) fac(ta) or similar.

⁹⁹ See note 93.

¹⁰⁰ Excavations directed by H. S. Gracie, who, with E. Price, made the stamps available for inspection and provided full details.

¹⁰¹ Clifford, Elsie, JRS xlv (1955), 68–72Google Scholar.

¹⁰³ BGAST lxxxix (1970), 40Google Scholar, where it is identified with Clifford ie.

¹⁰³ Ibid. 40, where it is identified with Clifford if.

¹⁰⁴ Ibid. 39.

¹⁰⁵ Ibid. 40.

¹⁰⁶ By the Kingscote Archaeological Association under the direction of E. J. Swain who provided details. The cube which is the subject of a note by M. Henig (Antiq. Journ., forthcoming) is dated to the second half of the third century when the cult of Sol Invictus was promoted by the emperors for propaganda purposes. Dr Henig cites as a parallel a coin of Victorinus, portraying the head of Sol with the legend INVICTVS (RIC, Victorinus, No. 96), and for the style of the bust on the cube, coin portraits of the emperor Aurelian (e.g. RIC, Aurelian, No. 395). Dr Henig thinks that the cube may have been used as an official seal. For the site, see p. 414.

¹⁰⁷ Information on this and the following two items from E. J. Swain and P. Arthur.

¹⁰⁸ By Helen Humphries in wh,ose possession it remains. Rubbing and details from W. J. Rodwell who made the object available for inspection. There seems to be no doubt as to the somewhat unexpected provenance of this find. The die does not appear to be matched in W. F. Grimes, Holt: The works depot of the Twentieth Legion at Castle Lyons.

¹⁰⁹ Mr M. Biddle directed the excavation for the (then) Ministry of Works (JRS xlix (1959), 131)Google Scholar. The graffito had been cut on fourand five-grooved lattice work. H. Sumner, Excavations in New Forest pottery sites (1927), 98, fig. XXVII, with block reused in JRS xv (1925), 242Google Scholar, fig. 107, No. 2, recorded analogous decoration from Rough Plece, Linwood, Hants. See Hants FCP, forthcoming.

¹¹⁰ Excavations for Winchester Excavation Committee were directed by M. Biddle. For the site see Antiq. Journ. xlviii (1968), 280–4Google Scholar. Information from Katherine Barclay who submitted this and the following three items for inspection.

¹¹¹ Antiq. Journ. 1 (1970), 311–21Google Scholar.

¹¹² Antiq. Journ. lv (1975). 295–303Google Scholar.

¹¹³ In (b) the fourth letter may be F, but certainly intended for E. The nomen Nicerius is formed from the common cognomen Nicerus, which is Greek in origin and of which 30 examples are attested at Rome in CIL vi. One urna contained 13-13 litres, and eight urnae, 105 litres, would be within about two per cent of the estimated volume of the vessel. For m(ensuravit), compare the suggested expansion m(ensura) in No. 44. For Roman measures of capacity see Hultsch, F., Griechische und romischc Melrologie, ed. 2 (Berlin 1882), especially p. 704, Table XIGoogle Scholar.

¹¹⁴ During rescue excavations by the East Herts. Archaeological Society Excavation Group directed by C. Partridge who submitted the sherds for inspection.

¹¹⁵ Referred to as ‘A 10 Cemetery B’, to distinguish it from the nearby cemetery at Skeleton Green.

¹¹⁶ See ILS index xvi, 834, nine examples.

¹¹⁷ Submitted by Professor S. S. Frere.

¹¹⁸ Submitted by Professor S. S. Frere.

¹¹⁹ Detected in 1971 in the store of Richborough Museum by Dr K. Greene. In Britannia v (1974), 393Google Scholar, fig. 6 he discusses the type of pottery and this so-called ‘monogrammatic’ cross on coins. He notes that a Christian ring was also found in 1931 in clearing the surface in the same area (Cunliffe, , Richborough v, 98Google Scholar, No. 160). R.P.W.

¹²⁰ Excavations directed by Christina Colyer for the Lincoln Archaeological Trust. For the site see above p. 390. The stamps were made available for inspection by Jenny Mann who also supplied casts and squeezes.

¹²¹ Excavations were conducted by the Department of Urban Archaeology of the Guildhall Museum (now Museum of London) under the direction of B. Hobley. Information on this and No. 95 from M. Rhodes, Assistant Archaeology Officer. For the site see Britannia vi (1975), 268Google Scholar.

¹²² For the cognomen Dubitatus, see CIL vii 495 (Legionary shield-boss recovered from the Tyne) and JRS xlix (1959), No. 21Google Scholar.

¹²³ Excavated by the Southwark and Lambeth Archaeological Society for the Southwark Archaeological Excavation Committee under the direction of G. Dawson who provided details and a drawing.

¹²⁴ Excavated by the Middle Nene Archaeological Group for the Nene Valley Research Committee under the direction of J. A. Hadman and S. G. Upex. See Current Archaeology No. 56 (May 1976) (published April 1977), p. 274 with photograph. Full information from Mr Hadman.

¹²⁵ Information from T. Ambrose.

¹²⁶ Rescue excavations directed by J. P. Greene for the Bath Excavation Committee. Information from T. Ambrose.

¹²⁷ During excavation for the University of Bristol directed by Dr (now Professor) K. Branigan who has sent the sherd and details; for the site see Britannia vii (1976), 357Google Scholar. The sherd lay in the debris of the last period of the stone building abandoned probably c. 370–80. For a similar Chi-Rho from Richborough see No. 92 (above), R.P.W.

¹²⁸ Excavation by P. J. Leach for the Committee for Rescue Archaeology in Avon, Gloucestershire and Somerset. For the site see Britannia vi (1976), 357Google Scholar. Information from W. J. Rodwell.

¹²⁹ The first letter may be an abandoned trial attempt. The name Verina is common (125 examples in CIL) and occurs twice in British inscriptions (RIB 447, JRS li (1961), 194Google Scholar, No. 12). Vrilucolus, however, is unmatched although several Celtic names begin with the prefix VRI.

¹³⁰ Information from T. Ambrose. The sherd is in the collection of the Roman Baths and Museum.

¹³¹ When found the tank contained a mass of waste lead and iron objects. Subsequent excavations for the Department of the Environment were directed by S. E. West who provided full details; see Britannia iii (1972), 330Google Scholar. and East Anglian Archaeology Report No. 3 (1976), 63–125Google Scholar. The tank was purchased bv Ioswich Boroueh Council in 1972 and is now on disolav in Inswich Museum, where facilities for inspection were provided by the curator, Patricia Butler, and C. J. Balkwill. Assistant Keener of Archaeology. For the Christian lead tank found previously at Icklingham, now in the British Museum, see JRS xxxiii (1942), 80Google Scholar, No. 12, pl. VIII.

¹³² By T. Walls; full information and drawings of the counters, now in Bourne Hall Museum, Ewell, were supplied by J. Cotton.

¹³³ Sent by Mrs K. F. Hartley, who directed the excavation for the (then) Ministry of Public Buildings and Works and Birmingham City Museum. The kiln was of late thirdto fourthcentury date; see Britannia i (1970), 286Google Scholar. R.P.W.

¹³⁴ Excavated by I. M. Stead for the Department of the Environment. Information on this and the following item provided by Valery Rigby. M.W.CH.

¹³⁵ Mr J. R. Perrin sent it for the York Archaeological Trust. For ol(l)a on a cooking-pot from Corbridge see JRS xxxviii (1948), 104Google Scholar, No. 25. R.P.W.

¹³⁸ During excavations directed by W. H. Manning for University College, Cardiff, and the Department of the Environment. Dr Manning made this and the following item available for inspection. For the excavations of 1973 see Britannia v (1974), 401Google Scholar.

¹³⁷ For the excavation of 1970 see Britannia ii (1971), 246Google Scholar.

¹³⁸ ‘In hoc vince’ (Rufinus, translating Eusebius, Hist. Eccl. ix.9.3). The letters are cut on top of the stone, not on its squared face (like a Roman ‘centurial’ stone), and thus after it came to lie in the field. The E, and probably the cs, are of non-Roman type. The stone was noticed by Mr S. Keay.

¹³⁹ The stone was noticed by Dr M. Roxan. The wall contains stones from the Milecastle gateways, and a building-stone, 0·32 by 0·18 m, with a well-cut recessed ansated panel, not inscribed.

¹⁴⁰ Mr G. C. Boon sent details.

¹⁴¹ Information from R.S.O.T. This amends JRS xlvii (1957), 234Google Scholar item (d) and RIB 1677, which in error assign it to the Museum of Antiquities, Newcastle upon Tyne. R.P.W.

¹⁴² By Professor Jocelyn M. C. Toynbee and R.P.W. She noted (to R.P.W. 10 Nov. 1976) that ‘the two little rolls of hair, one on each side beneath the ear, both ears being exposed, suggest the time of Severus Alexander’. In contrast the other person seems to have been conservative in her hairstyle.

¹⁴¹ R. P. Wright, ‘A Roman veterinary physician from the Thames Valley’ (see p. 279).

¹⁴⁴ JRS xv (1925), 250Google Scholar, No. 18. The object, untraced since 1940, and now on loan, was made available by Miss F. Gale for the Winchester Research Unit. The first name ends in the Celtic masculine nominative. The second, with the top of initial letter lost, may have been TER[TIVS, or by analogy TER[TIOS. See Hants FCP, forthcoming.

¹⁴⁵ See Cassianus, iie institutis coenobiorum 3,7. The phrase ‘relying on you’ is here withdrawn.

¹⁴⁶ Suggested by Mr KXS. Painter (in a lecture on 21 May 1977). For a comparable phrase see Jones, A. E., Current Archaeology No. 55 (1976, publ. 1977), 255Google Scholar.

¹⁴⁷ Mr K. S. Painter made it available for study. See also his The Water Newton Early Christian Silver (British Museum, 1977), p. 17Google Scholar, No. 12, pl. 12, fig. 8, reading ANICILLA, and J. P. C. Kent and K. S. Painter, Wealfh of the Roman World, A.D. 300–700 (1977), 32, No. 38. Desplte the damage at the left end the arrangement of the letters on either side of the vertical axis leaves insufficient space for a wide letter before the A. The first stroke is not part of a v but an I tilted backwards and topped by a V-shaped serif which extends to the left. This can be confirmed by viewing the embossed lettering from the verso.

The reading [T]VA ANCILLA is here withdrawn. The name Iamcilla seems to be unmatched, and is probably Celtic. The stem occurs in the nomen lamius (CIL ii, 767, Lusitania), and in the cognomina Iamus (CIL x, 6576, Velitrae) (also found in its Greek form) or Iamianus (CIL vi, 2410 B8). For the suffix Holder, Altkelt. Sprachsch. lists 23 examples of -c-illus.

¹⁴⁸ Dr D. C. A. Shotter and Mr J. G. Randall, of Lancaster University, consider that the theme is Dionysiac and that the text should be restored as TR]IVMP[HVS, presumably repeated two further times round the rim of the vessel. In 1975 R.P.W. (to R.S.O.T.) quoted the quintuple TRIVMPE of the carmen Arvale (ILS 5039, CIL vi 2104) and deemed it unacceptable. First, the hymn was addressed to Dea Dia, and not to Dionysus. Secondly, the surviving letters are carefully cut with serifs, and restoration of the forward-sloplng first extant letter as 1, preceded by R, to read TR]IVMP[HVS would cause the tail of the R to implnge upon the supposed 1. In contrast, in the restoration put forward in Britannia loc. cit., the L with vertical stroke inclined to the right would leave space for o and the foot of L, to read O]LVMP[VS.

¹⁴⁹ R.P.W. had the advantage of discussing this re-interpretation with Emerita Professor J. M. C. Toynbee, who referred him to C. Robert. The iconography is set out in full in C. Robert, Die antiken Sarkophag-Reliefs in, ii, pp. 242–67 (Berlin, 1904). For Rhea with tambourine see No. 196, pl. LXIII, No. 207, pl. LXVII and No. 208, pl. LXVIH; for Olumpus No. 201, pl. LXV, Nos. 208, 209, pl. LXVIH; and for the popularity of this theme in the Roman Emplre, pp. 243–4. It is to be hoped that a bowl with the figures and names more fully preserved will be found one day to confirm this reconstruction.