^{page 127 note 1} For the use of the terms in classical times, the reader should consult Lewis and Short's Latin Dictionary and Pauly-Wissowa, Realencyclopädie.

^{page 127 note 2} e.g. Carcopino, J., Daily Life in Ancient Rome, ed. Rowell, H. T. (1946), pp. 231–44.Google Scholar

^{page 127 note 3} Recherches à Salone, ii (1933), p. 43.Google Scholar

^{page 127 note 4} Bonner Jahrbücher, cxlv (1940), 35 and passim.Google Scholar

^{page 128 note 1} Archaeologia, lxxviii (1928), 111–218.Google Scholar

^{page 128 note 2} Chester Archaeological Journal (hereafter C.A.J.) xxviii (1929), 218–19, and xxix (1932), 66–8 and plan.

^{page 128 note 3} Ibid, xxix (1932), 5–40 and pls. I-XVI (plan on pl. XVI).

^{page 128 note 4} Determined, of course, by the curving outline of the amphitheatre.

^{page 130 note 1} Strictly, the Chester and North Wales Architectural, Archaeological and Historic Society, but the shorter title has passed into popular usage.

^{page 130 note 2} The history of the financial and administrative negotiations leading up to the final excavation of the amphitheatre can be traced in successive annual reports of the Society's Council as published in C.A.J. xxx (1933) to lvii (1970–1). A fuller account, with details of the personalities involved, appeared in the Cheshire Observer for 16th November 1957; this was based on notes provided by Mr. D. A. Nicholls who kindly made his files available to me.

^{page 131 note 1} The late Mr. T. Keith-Hill acted swiftly, in consultation with Alderman P. H. Lawson, to purchase the property from the Anglo-American Oil Company; there was some initial resentment but eventually their action was approved and, with Mr. C. W. Baty, they were appointed the first trustees of St. John's House.

^{page 131 note 2} C.A.J. xxxv (1945), 177; the St. John's House Account has figured in most subsequent volumes until the latest, lviii (1975), 137, 144.Google Scholar

^{page 134 note 1} See C.A.J. xxxviii (1951), 1–38; xxxix (1952), 21–8; xl (1953), 1–23.Google Scholar

^{page 134 note 2} With the exception of the legionary bath building in the praetentura which (as at Inchtuthil) seems to have been built in stone from the outset and which, on epigraphic evidence (R.I.B. 463), can probably be dated to A.D. 79.

^{page 135 note 1} C.A.J. xxxvi, pt. ii (1948), 104–5 and fig. 17.Google Scholar

^{page 138 note 1} For helpful descriptions of the various types of earthmoving machinery, see Petch, D. F., ‘Earthmoving machines and their employment on archaeological excavations’ (C.A.J. IV (1968), 15–28Google Scholar) and Pryor, Francis, Earthmoving on Open Archaeological Sites (Nene Valley Archaeological Handbook 1, 1974).Google Scholar

^{page 140 note 1} Cf. Richmond, I. A., ‘Roman Timber Building’ in Jope, E. M. (ed.), Studies in Building History (1961), p. 25 and fig. 1. 3.Google Scholar

^{page 140 note 2} The frequently cited passage from Pliny (N.H. xxxvi, 116–20) relates how Curio in 52 B.C. built two adjoining wooden theatres which could be rotated on pivots to form an amphitheatre. For a brief discussion of the structural problems involved, see Friedländer, L., Roman Life and. Manners under the Early Empire (Eng.translation of Sittengeschichte Roms, 7th edn.), 1913, ii, p. 43 and note in iv, pp. 511–12. Dio also records (xliii,22) that Caesar built a wooden amphitheatre in Rome as part of his triumph of 46 B.C.Google Scholar

^{page 140 note 3} C.I.L. v, 2, 7636: … in munus gladiatorium et saepta lignea; C.I.L. vi, 2059, 29 ff.: maenianum summum (sc. of the Colosseum) in ligneis, indicating that the topmost gallery was timber-built.

^{page 140 note 4} e.g. the part-stone, part-timber amphitheatre depicted on Trajan's Column (Lehmann-Hartleben, K., Die Trajanssäule (1926), p. 139 and Taf. 46); see pl. XLV aGoogle Scholar

^{page 140 note 5} Xanten in its first period (Bonner Jahrbücher, cxlv (1940), 34 and Taf. 6; Cirencester in its first period (Antiq. Journ. xliv (1964), 17 f.); and Maumbury Rings, Dorchester (above, pp. 38 ff.).Google Scholar

^{page 141 note 1} R. Laur-Belart, Vindonissa, Lager und Vicus, 1935 (Römisch-Germanische Forschungen, x), pp. 68–74 and Taf. 24.

^{page 141 note 2} Ibid., p. 74.

^{page 141 note 3} Klima, L. and Vetters, H., Das Lageramphitheater von Carnuntum (R.L.O. xx, 1953).Google Scholar

^{page 141 note 4} Ibid., pp. 25–48, Abb. 48, 51 and Beilage I.

^{page 141 note 5} Ibid., p. 29 n. 62.

^{page 141 note 6} Ibid., pp. 30–1 and Beilage II.

^{page 142 note 1} Bonner Jahrbücher, cxlv (1940), 50–1, Taf. 6 and Taf. 9, Abb. 1.Google Scholar

^{page 142 note 2} Sargetia, v (1968), 121–52, figs. 2, 13.

^{page 142 note 3} I must express my sincere thanks to Mr. Sunter for the care and patience he devoted to this task, especially in our preliminary discussions, and we are both grateful to Mr. John Weaver, F.S.A., for his helpful comments onstructural problems involved in timber buildings.

^{page 142 note 4} Rather than the fifth, since no allowance is being made for entrances. The circumference of the fourth row is calculated as π(113+98) ft. or π(34.3+29.8) m., where 113 ft. (34 3 m.) and 98 ft. (29.8 m.) are the radii of the fourth row on the long and short axes respectively.

^{page 142 note 5} Vitruvius (v, 1, 2) recommended 2–2½ Roman feet for theatres.

^{page 142 note 6} Richmond, I. A., ‘Trajan's Army on Trajan's Column’, P.B.S.R. xiii (1935), 30–1 and fig. 12.Google Scholar

^{page 142 note 7} Collingwood, R. G. and Richmond, Ian, The Archaeology of Roman Britain (1969), pp. 117–18.Google Scholar

^{page 142 note 8} G. C. Boon, Isca: The Roman Legionary Fortress at Caerleon, Mon. (1972), pp. 98 ff., quoting G. A. Webster, The Roman Imperial Armies of the First and Second Centuries A.D. (1969), pp. 201–2.

^{page 143 note 1} Undersökningär i Roms topografi (Göteborg, 1929), pp. 3–56. I am much indebted to Mrs. Eva Wilson for providing me with a translated summary of the relevant passage, pp. 22–8.Google Scholar

^{page 143 note 2} Pauly-Wissowa, , R.E. iii (1899), 1773, s.vv. castrense amphitheatrum, gives slightly different dimensions but if anything makes the cavea narrower still.Google Scholar

^{page 143 note 3} Lundstrom, op. cit., fig. on p. 25.

^{page 143 note 4} L. matutinus is thought to have derived its name from the venationes which took place in the morning (Platner, S. B. and Ashby, T., A Topographical Dictionary of Ancient Rome (1929), p. 320).Google Scholar

^{page 143 note 5} Val. Max. ii, 3, 2 (… ex ludo C. Aurelii Scauri doctoribus gladiatorunt accersitis); one is reminded of the formalized, almost balletic, bayonet drill practised in the British army in the present century.

^{page 143 note 6} Pauly-Wissowa, loc. cit. sup.; Platner and Ashby, loc. cit. sup.

^{page 143 note 7} Boethius, A. and Ward-Perkins, J. B., Etruscan and Roman Architecture (1970), pp. 270–1.Google Scholar

^{page 144 note 1} C.A.J. xxix (1932), II-19, pls. III, IV, VI, VII, XVI, and figs, 1 and 2.

^{page 146 note 1} C.A.J. xxix (1932), 16 ff. and frontispiece.

^{page 147 note 1} The rough masonry of the Caerleon arena wall was similarly rendered; see Archaeologia, lxxviii (1928), 118.Google Scholar

^{page 147 note 2} For 1930–1, cf. C.A.J. xxix (1932), 16, fig. 2 and pl. VII, 1 and 3.Google Scholar

^{page 148 note 1} For coping-stones at Caerleon, see Archaeologia, lxxviii (1928), 118 and fig. 3.Google Scholar

^{page 149 note 1} Archaeologia, lxxviii (1928), 114–15.Google Scholar

^{page 150 note 1} C.A.J. xxix (1932), 20–1, frontispiece and pl. VI.Google Scholar

^{page 150 note 2} It is perhaps hardly necessary to say that the Latin (h)arena had as its primary meaning ‘sand’, and was only later transferred to rriean the open space in the amphitheatre.

^{page 151 note 1} C.A.J. xxix (1936), pls. XIII and XVI.Google Scholar

^{page 151 note 2} Ibid. 19, 29, 36–7, and pl. x, 20–3.

^{page 151 note 3} Ibid. 20–1.

^{page 152 note 1} Archaeologia, lxxviii (1928), pl. XXII, 1.Google Scholar

^{page 152 note 2} Ibid. 153.

^{page 152 note 3} Ibid. pl. xx.

^{page 152 note 4} Ibid. 114–15, pl. xx and fig. 1.

^{page 156 note 1} Antiq.Journ. xliv (1964), 17–18 and xlvii (1967), 185–8.Google Scholar

^{page 156 note 2} Carmarthen Antiquary, vii (1971), 58–63.Google Scholar

^{page 156 note 3} Lehner, H., Vetera (Römisch-Germanische Forschungen, iv, 1930), pp. 68–71 and Abb. 50, 51.Google Scholar

^{page 157 note 1} C.A.J. xxviii (1929), 218, and xxix (1932), 10–11 and 66–8 (Site 1).Google Scholar

^{page 158 note 1} C.A.J. xxix (1932), 11–13 (Site 2 at the east end), 14–17 (Site 5 at the west end), pls. XII, XIII and XVI.Google Scholar

^{page 158 note 2} Ibid., pls. IV, 1 and V, 1.

^{page 158 note 3} C.A.J. xxxvi (1948), 103 (Site 9) and 104 (Site 10).Google Scholar

^{page 159 note 1} C.A.J. xxix (1932), 15, pls. v, 2, XII, and XVI.Google Scholar

^{page 161 note 1} C.A.J. xxix (1932), 13 and fig. 1.Google Scholar

^{page 161 note 2} Ignoring for the moment the width of the concentric wall and the area occupied by the various entrances.

^{page 162 note 1} Again ignoring the space required for the entrances and concentric wall.

^{page 162 note 2} No allowance is made for the Nemeseum, since it was excavated to the natural rock and would have produced some excess soil for disposal.

^{page 162 note 3} Archaeologia, lxxviii (1928), 115–17 and fig. 2.Google Scholar

^{page 162 note 4} Boon, G. C., Isca, the Roman Legionary Fortress at Caerleon, Mon. (1972), pp. 93–6.Google Scholar

^{page 163 note 1} Lehman-Hartleben, , Die Trajanssäule (1926), Taf. 46.Google Scholar

^{page 164 note 2} Archaeologia, lxxviii (1928), 117.Google Scholar

^{page 164 note 3} Mr. Sunter, in his reconstruction of the stone amphi-theatre (pp. 222–30 and figs. 45–9), argues for a timber-framed substructure over the inner width of the cavea in order to carry the seating.

^{page 166 note 1} C.A.J. xxix (1932), 14 and pl. xvi (Site 4).Google Scholar

^{page 166 note 2} Ibid., pl. VII, fig. 4 (sill) and pl. XIV.

^{page 167 note 1} Mr. Sunter, however, would offer a different interpretation (see below, p. 234).

^{page 169 note 1} Archaeologia, lxxviii (1928), 158–9 and fig. 12.Google Scholar

^{page 169 note 2} Ibid. 119–20 and fig. 4.

^{page 169 note 3} Klima, L. and Vetters, H., Das Lageramphitheater von Carnuntum (R.L.O. xx, 1953), pp. 40–2Google Scholar and Abb. 57, 61, 62. For the altars, Ibid. 55–9 and Abb. 75.

^{page 169 note 4} Conveniently listed in Pauly R.E. xvi (1935), 2361 (s.v. Nemesis).Google Scholar

^{page 170 note 1} C.A.J. xxix (1932), 11–12, pls. 11, XIII and XVI (= Newstead's Site 2).Google Scholar

^{page 170 note 2} Ibid. 14–15, pls. V, XII and XVI (= Site 5).

^{page 171 note 1} For a different interpretation of these features, see the section contributed by Mr. Nigel Sunter (below, p. 234).

^{page 173 note 1} Archaeologia, lxxviii (1928), 121 and fig. 1.Google Scholar

^{page 173 note 2} Archaeologia, lxxviii (1928), 135–43 and figs. 9–11.Google Scholar

^{page 173 note 3} C.A.J. xxix (1932), 11–12 and pls. II, XIII, and XVI.Google Scholar

^{page 173 note 4} Marked as ‘1930 trench’ on fig. 18, running at right angles to the north passage wall.

^{page 175 note 1} Mr. Sunter would interpret these features somewhat differently (see below, p. 233).

^{page 176 note 1} Mr. Sunter suggests a somewhat different reconstruction of the east entrance: see pp. 233–4 and fig. 50.

^{page 176 note 2} At Caerleon, similar flights of steps at axial entrances D and H differed in width, as if, the excavators suggested, to permit a more comfortable approach for the occupants of the tribunalia on the one hand and, on the other, better control of the general mass of spectators using these entrances (Archaeologia, lxxviii (1928), 135 and 139).Google Scholar

^{page 177 note 1} So far as it is legitimate to use the term with its judicial connotations; there are, however, one or two classical references in the sense of boxes (cf. Francovich, Geza de, Il Palatium di Teodorico a Ravenna (Rome, 1970), pp. 18–24).Google Scholar

^{page 177 note 2} Ibid., figs. 22–4.

^{page 177 note 3} Delbrueck, R., Die Consulardiptychen und verwandte Denkmäler (1929), pp. 218 ff. and Taf. 56.Google Scholar

^{page 177 note 4} Ibid., pp. 223 ff. and Taf. 58.

^{page 177 note 5} Ibid., Taf. 9–12, 20^v and 21^v; for a general discussion of tribunal, p. 80.

^{page 180 note 1} It is interesting to note that the first discovery of traces of the amphitheatre in 1929 (see above, p. 157 and fig. 1) fell in approximately the right position for the entrance nearest the southern main entrance on its west side, and that there were possible traces of the rubble foundation of one of the side-walls (fig. 11).

^{page 180 note 2} C.A.J. xxix (1932), 14–15, pls. V, XII, and XVI (Site 5).Google Scholar

^{page 181 note 1} Amphitheatre plans can be very irregular. Cf. the civil amphitheatre at Carnuntum (R.L.O. xvi (1926), Abb. 23 on p. 71).Google Scholar

^{page 182 note 1} The most recent statement of these events is Frere, S. S., Britannia (1967)Google Scholar, p. 117 (the more speculative account, Ibid., n. 2 seems excessively complicated.)

^{page 182 note 2} C.A.J. xxxviii (1951), 18 ff.Google Scholar

^{page 182 note 3} Thompson, F. H., Roman Cheshire (1965), p. 29.Google Scholar

^{page 182 note 4} Ibid., p. 13; Frere suggests as late as A.D. 383 under Magnus Maximus (Britannia, p. 361).

^{page 183 note 1} The classification is by Dyggve, Ejnar, in Recherches à Salone, ii (1933), 33–150 (L'Amphithéatre), and especially pp. 38–41. Apart from its detailed description of the amphitheatre at Salona, the work contains useful general comments on amphitheatre architecture and even a distribution map (fig. 14), based on Friedländer's list (Appendix xxxvi in Sittengeschichte Roms, iv). But the definitive work on Roman amphitheatres still remains to be written.Google Scholar

^{page 183 note 2} R.L.O. xx (1953), Beilage 1.Google Scholar

^{page 183 note 3} Szilágyi, J., Aquincum (1956), Beilage 1.Google Scholar

^{page 183 note 4} Boon, G. C., op. cit. (1972), n. 337 on p. 137.Google Scholar

^{page 183 note 5} Or, in Mr. Sunter's reconstruction (p. 231), on stone with a timber substructure for the inner part of the seating (fig. 53).

^{page 183 note 6} Sargetia, v (1968), fig. 13 on p. 138.Google Scholar

^{page 183 note 7} Bonner Jahrbücher, cxlv (1940), Taf. 9, 1.Google Scholar

^{page 184 note 1} J.R.S. lvii (1967), 203, pl. xvii, 1.Google Scholar

^{page 184 note 2} For dedications to Nemesis in amphitheatres see R.I.B. 323, a defixio (Caerleon); C.I.L. v, 17 (Pola); C.I.L. iii, 10439, 10442, 10446 (Aquincum), 11121 (Carnuntum). For ex visu in this cult see C.I.L. iii, 813, Pais 167 (both at Aquileia). For the cult of Nemesis, see C. B. Pascal, ‘The cults of Cisalpine Gaul’, Collection Latomus, lxxv, 38.

^{page 184 note 3} J.R.S. lviii (1968), 207, no. 8. The letters in 1. 1 are 19 mm., in 1. 2 65 mm., in 1. 3 64 mm. high.Google Scholar

^{page 184 note 4} J.R.S. lviii (1968), 208, no. 11.Google Scholar

^{page 184 note 5} For the comparable name Buricus see C.I.L. viii, 11400, xii, 2525.

^{page 184 note 6} J.R.S. lviii (1968), 208, no. 9.Google Scholar

^{page 186 note 1} For a centurial sign resembling a z see R.I.B. 1709 (Chesterholm).

^{page 186 note 2} J.R.S. lvii (1967), 203Google Scholar, no 7. For the discovery of other coping-stones see C.A.J. xxix (1932), 21, pl. 111, 3.Google Scholar

^{page 186 note 3} There is a partial leaf-stop after serano but the marks at the beginning and end of the phrase seem to be casual rather than intentional. The o in locvs was at first omitted; after its insertion a sloping foot was added to L.The cognomen Seranus is well attested in C.I.L. xii, xiii.

^{page 186 note 4} Places (loca) were assigned to the Arval Brethren in the Flavian Amphitheatre in A.D. 80 (C.I.L. vi, 32363,I.L.S. 5049). For Aries see C.I.L. xii, 714, 3; for Nîmes C.I.L. xii, 3316 (I.L.S. 5656); for Pola C.I.L. v, 86; for Syracuse C.I.L. x, 7130.

^{page 186 note 5} First published in J.R.S. lvii (1967), 203Google Scholar, no. 6 with the reading MET FEL. This has now been amended by W., R.P., Brit, iii (1972), 363.Google Scholar

^{page 186 note 6} J.R.S. lviii (1968), 208, no. 10.Google Scholar

^{page 186 note 7} Drawn by R.P.W. in 1970 in situ; Brit, ii (1971), 291Google Scholar, no. 8. For a similar stone see J.R.S. lviii (1968), 208, no. 12.Google Scholar

^{page 186 note 8} J.R.S. lviii (1968), 208, no. 12.Google Scholar

^{page 186 note 9} Brit. iii (1972), 355, no. 18.Google Scholar

^{page 187 note 1} C.A.J. xxix (1932), 25Google Scholar, fig. 3, reading IIRI Eri; ORI as drawn by W., R.P. (Brit, iii (1972), 355, no. 17). Orus without an initial aspirate seems to be the Latin form of the Greek name Oros, borne by a grammarian of the fifth century B.C. and a doctor from Mende quoted by Galen.Google Scholar

^{page 187 note 2} Brit. ii (1971), 294, no. 27.Google Scholar

^{page 187 note 3} J.R.S. lviii (1968), 213, no. 72.Google Scholar

^{page 188 note 1} C.A.J. xxix (1932), 19, 36–7.Google Scholar

^{page 188 note 2} C.A.J. xxxvi (1948), 100–2.Google Scholar

^{page 189 note 1} I am grateful to Miss G. Lloyd Morgan for her comments on some of these (noted as G.Ll.M.)

^{page 189 note 2} Archaeologia, lxxx (1930), 43 and fig. 3.Google Scholar

^{page 189 note 3} Antiq. J. xliii (1963), 289–90, figs. 1 and 2.Google Scholar

^{page 189 note 4} Ibid, lv (1975), 41–61.

^{page 190 note 1} Wild, J. P., ‘Button-and-Loop Fasteners in the Roman Provinces’, Britannia, i (1970), 137–55.CrossRefGoogle Scholar

^{page 190 note 2} Ibid. 141 and fig. 1.

^{page 192 note 1} Cf. Thompson, F. H., Roman Cheshire (1965), fig. 20, no. 1, for a folding example from Wilderspool.Google Scholar

^{page 198 note 1} I am indebted to Miss Dorothy Charlesworth, F.S.A., for her report on nos. 3–11.

^{page 200 note 1} The Samian has been examined partly by Joanna Morris and partly by B. R. Hartley, F.S.A., and I am much indebted to them both for their reports. The following references are used:

O: Oswald, F., Index of Figure-types on Terra Sigillata (1936–1937).Google Scholar

Knorr 1910: R. Knorr, Terra Sigillata von Rottenburg.

Knorr 1919: R. Knorr, Töpfer und Fabriken verzierter Terra-Sigillata des ersten Jahrhunderts.

Knorr 1952: R. Knorr, Terra-sigillata-gefässe des ersten Jahrhunderts mit Töpfernanten.

Atkinson 1914: D. Atkinson, ‘A Hoard of Samian Ware from Pompeii’, J.R.S. iv.

Camulodunum: C. F. C. Hawkes and M. R. Hull, Camulodunum (1947).

Fishbourne: Dannell, G. B., ‘The Samian Ware’, in Cunliffe, B. W., Excavations at Fishbourne, vol. ii (1971).Google Scholar

Holt: Grimes, W. F.‘The Works Depot of Legion XX at Holt (Castle Lyons), Denbighs.’, Y Cymmrodor, xli, 1930.Google Scholar

Verulamium: Hartley, B. R., ‘The Samian Ware’, in S. S. Frere, Verulamium Excavations (1972).Google Scholar

Stanfield 1936: Stanfield, J. A., ‘Unusual Forms of Terra Sigillata, Second Series’, Arch. J. xciii.Google Scholar

^{page 221 note 1} Since this paragraph was written, Mr. Weaver has drawn my attention to the Report of the Inspector of Ancient Monuments for the Year ending 31st March 1913, which contains much valuable information of this kind, especially the technical appendices I—III (pp. 54–117); unfortunately, this was never continued.

^{page 222 note 1} Boëthius, A. and Ward-Perkins, J. B., Etruscan andRoman Architecture (1970), p. 353.Google Scholar

^{page 222 note 2} Lehner, H., Vetera (Römisch-Germanische Forschungen, iv, 1930), pp. 68–71 and Abb. 50, 51.Google Scholar

^{page 222 note 3} See pp. 43–8.

^{page 222 note 4} Bonner Jahrbücher, cxlv (1940), 33–62, Taf. 6, 9 and 10.Google Scholar

^{page 222 note 5} Laur-Belart, R., Vindonissa, Lager und Vicus (Römisch-Germanische Forschungen x, 1935), pp. 68–74Google Scholar and Taf. 24; Fellmann, R., Führer durch das Amphitheater von Vindonissa (1952), Abb. 4.Google Scholar

^{page 223 note 1} Op. cit. in n. 4, p. 222, Taf. 6.

^{page 223 note 2} Grenier, A., Manuel d'archéologie gallo-romaine, iii, 2 (1958), pp. 734–41.Google Scholar

^{page 223 note 3} For a discussion of the setting-out of the amphitheatre, see the reconstruction of the stone amphitheatre, p. 230.

^{page 224 note 1} Grenier, op. cit., p. 567.

^{page 224 note 2} Klima, L. and Vetters, H., Das Lageramphitheater von Carnuntum, (R.L.O. xx, 1953).Google Scholar

^{page 224 note 3} Sargetia, v (1968), 121–52, and figs. 2, 13.Google Scholar

^{page 224 note 4} Laur-Belart, op. cit., pp. 68–74.

^{page 224 note 5} Jope, E. M. (ed.), Studies in Building History, (1961), pp. 15–26.Google Scholar

^{page 224 note 6} A greater height, 3 ft. 8 in. (1·1 m.), would be required only opposite the entrances to protect standing spectators since it is unlikely that a continuous gangway would have been provided in front of the seating.

^{page 224 note 7} Any excavation of the arena to a lower level would have necessitated an additional palisade set forward of the projecting ends of the sill-beams to retain the unexcavated ground in front of the cavea. This palisade, in order not to restrict unduly the view of the spectators, could have been no more than approx. 9 ft. 8 in. (2·8 m.) above lowered arena level (for discussion see p. 226). There would therefore appear on these grounds alone to be no practical gain in lowering the floor of the arena at the expense of providing an otherwise unnecessary palisade.

^{page 225 note 1} It is not known how deeply they were housed into the sill-beams.

^{page 226 note 1} For a discussion of the variation in the filling of the sill-beam trenches and the nature of the soil into which these were cut see p. 137 where this is explained as a result of site-levelling.

^{page 227 note 1} London Archaeologist, ii (1974), 157Google Scholar and now Trans. London and Middlesex Arch. Soc. xxv (1974), 122–8 and figs. 7–9.Google Scholar

^{page 227 note 2} London: London Archaeologist, ii (1974), 158Google Scholar; Dover: Arch. J. cxxvi (1970), 90–2.Google Scholar

^{page 227 note 3} Jope, op. cit., pp. 15–26.

^{page 227 note 4} I am grateful to Mr. Stuart Rigold, F.S.A., for this information.

^{page 228 note 1} I am most grateful to Mr. Dennis Avery, A.R.I.C.S., who has undertaken the task of assessing the total quantity of timber involved in the reconstruction. For the purposes of this exercise, the same construction has been assumed for every bay, and no adjustment has been made for entrances and access stairs to the seating.

^{page 228 note 2} I am most grateful to Mr. Burton, who stresses that his calculation is an ‘inspired guess’ which unknown local conditions could well have influenced to a large degree. The extensive range of between 500 and 1,500 H.ft. per acre is an indication in itself of the futility of attempting to give any accurate assessment of the acreage. It is interesting to note that by a different method the large double row of timbers around the front of the cavea at Maumbury is calculated to have required approx. 120 acres (47 ha) of forest within which to find suitable trees (see p. 39, and Nørlund, P., Trelleborg (Nordiske Fortidsminder iv, 1, 1948), pp. 175–8).Google Scholar

^{page 230 note 1} I would particularly like to thank Mr. Hugh Thompson, Mr. Stuart Rigold, and Mr. John Weaver for their help and constructive criticism during the preparation of the timber amphitheatre reconstruction.

^{page 230 note 2} Britannia, i (1970), 87–94.Google Scholar

^{page 231 note 1} Ibid. 86 and 90.

^{page 231 note 2} Dr. Ward-Perkins suggests that these amphitheatres could have dated to the late Flavian or Trajanic periods (Boëthius and Ward-Perkins, op. cit., p. 353).

^{page 231 note 3} Boon, G. C., Isca, The Roman Legionary Fortress at Caerleon, Mon. (1972), pp. 90–7 (hereafter referred to as Isca, 1972).Google Scholar

^{page 231 note 4} Boon, G. C., Isca, A Guide (1962), frontispiece.Google Scholar

^{page 231 note 5} I find plausible Mr. Boon's idea that the upper stage of the amphitheatre wall was completed in timber, but his evidence unconvincing (Isca, 1972, p. 95); nevertheless it is almost certain that, as Mr. Boon acknowledges, the architect would originally have intended the exterior to be completed in stone.

^{page 231 note 6} Grenier, op. cit., pp. 860–7.

^{page 231 note 7} Archaeologia, lxxxiv (1935), 213–61.Google Scholar

^{page 231 note 8} Grenier, op. cit., pp. 856–9.

^{page 231 note 9} Boëthius and Ward-Perkins, op. cit., p. 353. However, it is argued by Mr. D. S. Robertson that at Orange the roof over the pulpitum was self-supporting and that only twelve of the forty-two corbels behind the rear wall of the scaenae frons could possibly have carried masts which projected above roof level; Robertson, D. S., Greekand Roman Architecture (second edition, 1943), pp. 281–2. If Mr. Robertson's argument is right, at Les Bouchauds the paired rear walls of the scaenae frons could also have supported such a roof which might not therefore have required any additional support of masts.Google Scholar

^{page 232 note 1} Grenier, op. cit., pp. 808–14.

^{page 232 note 2} Note that the plan (Archaeologia, lxxxiv (1935), pl. LXVII) shows only those opposite the buttresses.Google Scholar

^{page 232 note 3} Boëthius and Ward-Perkins, op. cit., p. 224.

^{page 232 note 4} Ibid., p. 223.

^{page 233 note 1} Bonner Jahrbücher, cxlv (1940), Taf. 6, 9, and 10.Google Scholar

^{page 233 note 2} Laur-Belart, op.cit., Taf. 24; Fellmann, op. cit., Abb. 4.

^{page 233 note 3} See pp. 47–8.

^{page 233 note 4} At Caerleon (Isca, 1972, p. 94) a combination of posts and sill-beams is suggested.

^{page 233 note 5} Fellmann, op. cit., p. 6.

^{page 233 note 6} Grenier, op. cit., pp. 593–6.

^{page 233 note 7} The spacing of the ‘abandoned’ wall from minor entrance no. 4 is consistent with the spacing of the remaining excavated entrances one to the other. It is also worth noting that these inner walls are laid out not from the centre point of the arena, but in relation to the two ‘abandoned’ walls, a fact which would itself suggest that they are a secondary feature.

^{page 234 note 1} Alternatively there may have been a complete break in the façade wall; the foundations terminate either side of the entrance and in this respect it differs from the east entrance.

^{page 234 note 2} I am grateful to Mrs. Susan Walker for this suggestion.

^{page 234 note 3} Grenier, op. cit., p. 595.

^{page 234 note 4} Laur-Belart, op. cit., Taf. 24; Fellmann, op. cit., p. 6.

^{page 234 note 5} Grenier, op. cit., p. 653.

^{page 234 note 6} Bonner Jahrbüher cxlv (1940), Taf. 6.Google Scholar

^{page 234 note 7} Antiq. Journ. xliii (1963), 23–6.Google Scholar

^{page 236 note 1} I am indebted to Mr. Thompson who has devoted a considerable number of lunchtimes to and expended much energy on constructive discussions without which this reconstruction would not have been possible. I am also grateful to Mrs. Susan Walker for her suggestions and help.