¹ Lysons, S., Reliquiae Britannico-Romanae vol. iii (1817)Google Scholar [1819] drawings, not text; idem, Archaeologia xviii (1817), 203-21; ibid., xix (1821), 176-7. There exist several copies of a volume of typescript, Notes on the Roman Villa at Bignor, Sussex, compiled by George Herbert, it seems soon after 1927, which gives a useful critical summary of the villa down to that date. See also S. E. Winbolt and George Herbert, The Roman Villa at Bignor, Sussex, a guide-book first published in 1925 and re-issued in successive editions until replaced by a more ‘popular’ account in recent years. See also V.C.H. Sussex iii (1938), 22; JRS liii (1963), 155 f., and Britannia v (1974), 457.

² It was not possible in recent work to re-excavate and establish the position of the N.E. corner of the court-yard, but the whole of the alignment of the S. corridor was fixed and that of the N. corridor as far as a point opposite Room 13. On FIG. 2 the E. corridor is shown as 127 ft. long. Lysons's plan shows it as 120 ft. and his text as 133 ft. long. An idea of the somewhat haphazard planning of the original excavations can be gained from the correspondence published by Steer, F. W. (ed.), The Letters of John Hawkins and Samuel and Daniel Lysons 1812-1830 (West Sussex County Council, 1966).Google Scholar

³ For his account see Sussex Arch. Colls, lxvii (1926), 84-88.

⁴ Thanks must be expressed to the late Capt. Henry Tupper for his encouragement of the work, and to Miss M. G. Wilson, Miss H. Waugh and Mrs C. M. Bennett, as well as to the numerous other volunteers, without whose assistance little could have been achieved.

⁵ In Todd, M. (ed.), Studies in the Romano-British Villa (Leicester, 1978), 174.Google Scholar

⁶ The tondo is painted pinkish-yellow with darker pink or purple veins. This is surrounded by a green band with white speckles which is sunk 1 in. (0·025 m) into the wall. Beyond this the normal surface resumes with a circular white border carrying inward-pointing brown-painted triangles which suggest rays. Next comes an area of yellow with red veins on which is part of a large purple leaf. Beyond a purple border is a panel of pink with red veins, and this is bordered by a vertical band of black with white veins recessed 1 in. into the wall and edged each side by white bands bearing brown-black lines to indicate mouldings. Outside this is an area of pinkish-red with red veins, then a moulding represented in black and white paints, and next a yellow panel with red veins. Bordering this at the surviving edge is a black band recessed 1 in. and with its moulded edge turning back left at the bottom. The panel is further discussed in N. Davey and R. Ling, Wail-Painting in Roman Britain (Britannia Monograph No. 3, 1982), No. I.

⁷ Davey, N. in Britannia iii (1972), 265.Google Scholar

⁸ In Section X–Y, dug in 1958, this division of the buried soil had not been noticed.

⁹ This trench yielded a piece of shale flooring-flag—presumably from Room 55 or 56, the only rooms where this material is recorded: see FIG. 31.

¹⁰ Britannia ix (1978), 154, 165.

¹¹ In Munby, J. and Henig, M. (eds.), Roman Life and Art in Britain (BAR No. 41, 1977), 195–215Google Scholar; cf. also in Todd, M. (ed.), Studies in the Romano-British Villa (Leicester, 1978), 88 f.Google Scholar

¹² Both lead pipes are c. 2 in. (0·05 m) in external diameter and 1½ in. (0·04 m) in internal. The pipe visible in the centre of the base of the piscina is the inlet-pipe; the arrangements for the outlet are invisible, but presumably it drained a lead sink lying below the inlet. Both pipes appear to use the same aperture in the base of the piscina.

¹³ I am much indebted to Mr Mark Owen Jones of the Hydraulics Research Station, Wallingford, for discus-sing these questions with me. For a note on the water-supply, see Sussex Notes and Queries xv (1960), 192-6.

¹⁴ Corder, P. and Kirk, J. L., A Roman Villa at Langton near Malton, E. Yorkshire (Leeds, 1932), 48 f.Google Scholar

¹⁵ Applebaum calculated this as possibly as much as 800 acres (= 324 ha) for corn-growing in Finberg, H. P. R. (ed.), The Agrarian History of England and Wales (Cambridge 1972), 241; cf. Britannia vi (1975), 118-32.Google Scholar

¹⁶ Thanks are due to Dr C. E. King for much valuable criticism and advice.

¹⁷ Except No. 10, a coin exhibited by Mr T. Honeywood of Horsham in July 1855, on the occasion of the A.G.M. of the Sussex Archaeological Society in the Town Hall, Horsham (Sussex Arch. Colls, viii (1856), x). No further details are given, and the provenance must be open to question.

¹⁸ The probable reverse types are either an eagle or an altar (RIC v. 2 (Tetrici) 164, 168).

¹⁹ Probably RIC v. 2 (Tetrici) 270-4, legends SPES AVGG or SPES PVBLICA.

²⁰ Probably RIC vi (Londinium) 83-6 (?summer 307).

²¹ The coin is described as ‘worn’, but even so, it is unlikely to be from the same series as No. 16. If it was struck after 310, on the other hand, the smaller module would be normal; the ruler could then be either MAXIMINVS DAIA or Galerius (MAXIMIANVS).

²² An impossible reading, SARL, the mint-mark of Aries, is likely; the diameter is consistent with early issues from this mint. The possible date is therefore 313-5 (cf. RIC vii 14-16, 19-22, 35-41, 44-5).

²³ An improbable reading; coins with the legend SECVRITAS PVBLICA were struck only at Constantinople by Hanniballian. SECVRITAS REIPVBLICE is almost certain, probably with the mint-mark PTR or its variants (LRBC 25, 35, 41, 47).

²⁴ Probably an example of the coinage produced by Magnentius and Decentius in 351-3. The legend should be VICTORIAE DD NN AVG ET CAES(S) and the type two Victories facing each other, holding a shield inscribed VOT/V/MVL/X. The issue was struck, with variations, at several mints, including Amiens (LRBC 5–18), Lyons (LRBC 217–35), Trier (LRBC 56–61) and Aries (LRBC 437–42).

²⁵ Cf. Reece, R., Britannia iii (1972), 272 (fig. 1), 274-5Google Scholar.

* The objects marked with an asterisk were stolen from the Bignor Museum and are known only from the illustrations, here reproduced, which were made shortly after discovery. It will readily be appreciated, therefore, the interpretations offered may not be as secure and detailed as might otherwise be expected. The remaining objects were conserved by Mrs S. Pollard and drawn by Mrs A. Wilkins at the Institute of Archaeology at Oxford.

²⁶ At Fishbourne, for instance, 19 of the 21 examples found were stratified in contexts dated A.D. 43-75 (Hull, M. R., ‘The Brooches’ in Cunliffe, B. W., Excavations at Fishbourne 1961–1969 Vol. ii (Report of the Research Committee of the Society of Antiquaries of London No. XXVII, 1971), 100).Google Scholar

²⁷ Cf. Mackreth, D. F., ‘The Roman Brooches’ in Down, A., Chichester Excavations iii (Chichester, 1978), 279, fig. 10.26.7.Google Scholar

²⁸ Thanks are due to Mr A. Havercroft, Dr M. Henig, Mr D. Mackreth and Professor J. M. C. Toynbee, who gave valuable advice.

²⁹ The class corresponds with Camulodunum Types xiii and xiv, and with Type 27 in Ettlinger, E., Die römischen Fibeln in der Schweiz (Bern, 1973)Google Scholar. The most recent study is by E. Riha, Die römischen Fïbeln aus Augst und Kaiseraugst (Forschungen in Augst Band 3, Augst 1979), 108-9, and the proposed categories A, B, C correspond broadly with her Variante 4.8.1, 4.8.2, 4.8.3.

³⁰ Bagendon: Clifford, E. M., Bagendon: A Belgic Oppidum (Cambridge 1961), 176, no. 42, St Albans: Antiq. Journ. xlvii (1967), 290.Google Scholar

³¹ E.g. Colchester: Camulodunum, 319-20, nos. 113, 115; Antiq. Journ. xxii (1942), 61, fig. 1.7 (associated with Claudian pottery). Chichester: Down, A., Chichester Excavations iii (Chichester, 1978), 286Google Scholar, no. 252. Hod Hill: Brailsford, J. W., Hod Hill i: Antiquities from Hod Hill in the Durden Collection (London, 1962), 8, no. C.43. Longthorpe: Britannia v (1974), 44, no. 6.Google Scholar

³² Cf. Gonzenbach, V. V., ‘Tiberische Gürtel- und Schwertscheidenbeschläge mit figürlichen Reliefs’, Helvetia Antiqua: Festschrift Emil Vogt (Zurich, 1966), 183 ff.Google Scholar

³³ Matz, F., ‘Zum Axretiner Thiasos’, Festschrift für August Oxé (Darmstadt, 1938), 9 ffGoogle Scholar. A similar Maenad appears on an Arretine crater in the Ashmolean Museum (Brown, A. C., Catalogue of Italian Terra-Sigillata in the Ashmolean Museum (Oxford, 1968), 18Google Scholar, No. 46).

³⁴ Cf. Robertson, M., A History of Greek Art (Cambridge, 1975), 605 f.Google Scholar

³⁵ Cf. Furtwängler, A., Die Antiken Gemmen i (Berlin, 1900)Google Scholar, Taf. 38.16. Dancers, satyrs and centaurs, often painted on a white background and thus reminiscent of those on Athenian white-ground vases, are common subjects during this period (cf. Rumpf, A., Malerei und Zeichnung (Handbuch der Archäologie Band 4, München 1953), 175)Google Scholar.

³⁶ Strong, D., Greek and Roman Gold and Silver Plate (London, 1966), 177–8Google Scholar. Cf. Liversidge, J., Britain in the Roman Empire (London, 1968), 159, fig. 67.C (an example from Richborough)Google Scholar.

³⁷ Down, A., Chichester Excavations iii (Chichester, 1978), 310Google Scholar, fig. 10.43, and 177-8; id., Chichester Excavations iv (Chichester, 1979), 155, fig. 48.32.

³⁸ Robertson, A. S., Birrens (Blatobulgium) (Edinburgh, 1975), 128, fig. 43.1.Google Scholar

³⁹ Nos. 1-5 are probably of bone (certainty is clearly impossible in the case of the stolen objects), 6-8 of antler. Thanks are due to Dr A. MacGregor for assistance in distinguishing between the two materials.

¹⁰ Cf. Wild, J. P., Textile Manufacture in the Northern Roman Provinces (Cambridge, 1970), 73–4, 140-1 (Table o), fig. 63Google Scholar.

⁴¹ Crummy, N., ‘A Chronology of Romano-British Bone Pins’, Britannia x (1979), 157–63.CrossRefGoogle Scholar

⁴² The writer has benefitted from valuable discussions with Messrs A. Down, C. M. Green, M. Millett, R. P. Symonds and Dr C. J. Young. Mr D. Rudkin (Fishbourne Roman Palace), Miss E. Kelly (Worthing) and Mr D. Watts (Ashmolean, Oxford) kindly facilitated access to material in the museums named.

⁴³ 27 per cent measure less than 2 cm square, 49·5 per cent between 2 and 4 cm square, 19 per cent between 4 and 6 cm square, and only 4 5 per cent more than 6 cm square. The dating of the sherds was by Mr B. R. Hartley.

⁴¹ Apart from the early imports, the sources of the white or buff vessels are particularly difficult to identify. Many share a sandy fabric generally similar but not identical to that produced in the Wiggonholt kilns (Sussex Arch. Colls, cxii (1974), 130-2) and are probably of local manufacture.

⁴⁵ Arthur & Marsh, 62.

⁴⁶ ibid. 308-9.

⁴⁷ From a vessel similar to that illustrated ibid. fig. 2.3.2.

⁴⁸ A. C. Anderson, A Guide to Roman Fïne Wares (1980), 14-20 (Lower Rhineland Fabric 1).

⁴⁹ The argumentum e silentio is inevitably weak, but prima facie the proportion of this ware may be compared with that at the Chilgrove I villa, which, it is suggested, ceased to be used as domestic accommodation in the second half of the fourth century (Down, A., Chichester Excavations iv (Chichester 1979), 48–51, 194-7)Google Scholar.

⁵⁰ Two examples from the St Pancras cemetery were associated with Antonine samian (Down, A. & Rule, M., Chichester Excavations i (Chichester, 1971), 115 No. 235 B, 117 No. 250 A).Google Scholar

⁵¹ The fabric is generally similar to Fishbourne Fabric B.

⁵² Cf. Hodder, I., ‘The Distribution of Two Types of Romano-British Coarse Pottery in the West Sussex Region’, Sussex Arch. Colls, cxii (1974), 86–96.Google Scholar

⁵³ Lysons, S., Reliquiae Britannico-Romanae iii (1817) [1819]Google Scholar, pl. xxxi, fig. 10. It was illustrated and briefly discussed by Kendrick, T. D. (Sussex Notes and Queries vii (1938–1939), 244–5), and also in an anonymous note in Antiq. Journ. vii (1927), 521. In both these articles it was assigned to the late Roman period, but the parallels cited are not at all close.Google Scholar

⁵⁴ See Marshall, F. H., Catalogue of the Finger Rings, Greek, Etruscan and Roman in the Departments of Antiquities British Museum (London, 1907), xviii–xxi.Google Scholar

⁵⁵ In a letter to Lysons (published by Winbolt, S. E., Sussex County Magazine i No. 6 (May 1927), 272) Hawkins mentions ‘the discovery of a gold ring. We have purchased it of the fortunate finder. It weighs about one Guinea and a 7 shilling] piece’; see also F. W. Steer, op. cit. (note 2), 45. Its subsequent history is discussed by George Herbert, Notes on the Roman Villa at Bignor, Sussex (op. cit., note 1), 66-7.Google Scholar

⁵⁶ Sussex Arch. Colls, viii (1856), 292.

⁵⁷ Dimensions: hoop 24 mm (external diameter), 19 mm (internal); bezel 19 mm × 17 mm.

⁵⁸ 10 mm × 8 mm. cf. M. Henig, A Corpus of Roman Engraved Gemstones from British Sites (BAR 8, second edition 1978), 219, no. 278.

⁵⁹ Henkel, F., Die römischen Fingerringe der Rheinlande (Berlin, 1913), 40 and pl. xiv no. 274; 42-3 pl. xv nos. 288, 289Google Scholar. Beckmann, C., ‘Metallfingerringe der romischen Kaiserzeit im freien Germanien’, Saalburg Jahrbuch xxvi (1969), 34–5 nos. 664, 668, 669, 687 (Form 17a).Google Scholar

⁶⁰ Charlesworth, D., ‘Roman Jewellery found in Northumberland and Durham’, Arch. Ael. ⁴ xxxix (1961), 17 and 31 (no. 96), pl. IV. 13. See also 4 and 30 (no. 79), pl. iv.4, for rings from Backworth with pellets, apparently of second-century date.Google Scholar

⁸¹ Maaskant-Kleibrink, M., Catalogue of the Engraved Gems in the Royal Coin Cabinet, The Hague (The Hague, 1978), 326Google Scholar. For style and motif, see Zwierlein-Diehl, E., Die Antiken Gemmen des Kunsthistorischen Museums in Wien ii (Munich, 1979), 202Google Scholar no. 1479 (pl. 144), where our gem is cited for theme, but not for stylistic resemblance. The date proposed here is the third century.

⁶² Reliquiae pl. xxxii, 9 and 10. Found in Room 54. The body of the statuette was evidently not located. The caption says it was originally painted. If so, it is to my knowledge the only recorded instance in Britain of what must have been a common occurrence in antiquity. However we cannot be sure that what Lysons saw was not merely staining on the stone.

⁶³ R. E. M., and Wheeler, T. V., Excavation of the Prehistoric, Roman and Post-Roman Site in Lydney Park, Gloucestershire (Report of the Research Committee of the Society of Antiquaries of London No. ix, London 1932), 68Google Scholar, pl. xxiv a; cf. Toynbee, J. M. C., Art in Britain under the Romans (Oxford, 1964), 91.Google Scholar

⁶⁴ Chichester iv, 181-3, pl. xi; cf. Esperandieu, E., Recueil General des Bas-Reliefs, Statues et Bastes de la Germanie Romaine (Paris, 1931), 241 no. 375.Google Scholar

⁶⁵ Britannia v (1974), 242, pl. xx a.

⁶⁶ Antiq. Journ. lxi (1981).

⁶⁷ Morton, G. R. & Wingrove, J., ‘Constitution of Roman Bloomery Slags’, Journal of the Iron and Steel Institute 207 (1968), pp. 1554–64.Google Scholar