^{page 129 note 1} Collingwood, R. G. and Richmond, Ian, The Archaeology of Roman Britain (1969), p. 101Google Scholar.

^{page 130 note 1} Briefly reported in Current Archaeology, xxvi (May 1971), 67–71Google Scholar.

^{page 130 note 2} Brief notes of excavations at the east gate have appeared as follows: J.R.S. 1 (1960), 221; li (1961), 170; lii (1962), 167; liii (1963), 131; liv (1964), 159; lv (1965), 205; lvi (1966), 203; lvii (1967), 181; the excavations are discussed in J. B. Whitwell, Roman Lincolnshire (1970), pp. 21 et seqq.

^{page 132 note 1} Arch. Journ. ciii (1946), 33–4 and references.

^{page 132 note 2} Ibid. 34, pl. VII c.

^{page 132 note 3} The writer is indebted to Mr. D. F. Petch for providing notes and drawings of his excavation.

^{page 132 note 4} Cf. V. E. Nash-Williams, The Roman Frontier in Wales (2nd edn., 1969), p. 35.

^{page 132 note 5} It should be pointed out that this evidence is at variance with that obtained from ditch 1 within the north tower of the gate, for which see pp. 135, 149.

^{page 133 note 1} P. Corder, A Romano-British Pottery Kiln on the Lincoln Racecourse, 1950 (Nottingham University publication).

^{page 138 note 1} It has now also been recognized on the northern defences at East Bight, where the early colonia wall construction was identical to that encountered at the east gate. Cf. J.R.S. lv (1965), 205Google Scholar; lvi (1966), 203; lvii (1967), 181, for brief notes; full report forthcoming.

^{page 138 note 2} The north and south gates of the legionary fortress at Vindonissa show a comparable casing in stone (Claudian) of the earlier timber defences (Tiberian). Cf. Belart, R. Laar, ‘Vindonissa, Lager und Vicus’, Römisch-germanische Forschungen, xix (1961), 24–8Google Scholar and Taf. 8.

^{page 140 note 1} These details of the behaviour of the early colonia town wall were in every way echoed at East Bight (p. 138, n. 1).

^{page 140 note 2} See Arch. Journ. ciii (1946), pl. viiib, for a double phallus built into the cottage east of Newport Arch, Mrs. C. Wilson has told the writer of a further phallus found on the bottom course of the town wall at East Bight.

^{page 141 note 1} The diameter of the Holme Pierrepont wheel is 33 in. (Transactions of the Thoroton Society, lxxii (1968), 23Google Scholar), and that of the Newstead wheel is 41 in. (J. Curie, A Roman Frontier Post and its People (1911), p. 294).

^{page 141 note 2} Arch. Journ. ciii (1946), 34Google Scholar and n. 27.

^{page 144 note 1} S. S. Frere, Britannia (1967), p. 71, n. 2; Whit-well, J. B., Roman Lincolnshire (1970), p. 19Google Scholar.

^{page 144 note 2} J.R.S. xliv (1954), 84Google Scholar, fig. 8.

^{page 144 note 3} The post-pits at the east gate of Inchtuthil are at 9-ft. spacings.

^{page 149 note 1} This width of roadway is similar to the east gate at Inchtuthil (J.R.S. xliv (1954), fig. 8 on p. 84) and the north gate at Longthorpe (J.R.S. lviii (1968), fig. 13 on p. 190).

^{page 149 note 2} Frere, S. S., Britannia (1967), p. 70Google Scholar.

^{page 149 note 3} Longthorpe, and Newton, in J.R.S. lv (1965), 74–6Google Scholar, figs. 1 and 2, pls. ix, and x, 1; Frere, S. S., Britannia (1967), pp. 70–1Google Scholar; Whitwell, J. B., Roman Lincolnshire (1970), pp. 11–12Google Scholar.

^{page 149 note 4} J.R.S. xxxix (1949), 64–6Google Scholar.

^{page 149 note 5} A scrap of decorated samian from the first occupation surface (fig. 19, no. 11) cannot be earlier than c. a.d. 70 (p. 173). It should probably be related to the second phase of legionary activity.

^{page 149 note 6} Arch. Journ. cxvii (1960), 40–70Google Scholar.

^{page 149 note 7} The evidence of early colonia activities in excavations at the east gate and in East Bight tends to support the late first-century date for the foundation of the colonia suggested by the late Professor Richmond, I. A. in Arch. Journ. ciii (1946), 29Google Scholar from the Mainz inscription.

^{page 150 note 1} By Miss C. Colyer, following excavations in 1968 directed by the writer. The partial excavation of the inner lip of the ditch at The Park in 1968 reached only the eighteenth-century filling recorded by the topographers; Arch. Journ. ciii (1946), 40, 74Google Scholar.

^{page 150 note 2} It should be admitted that they were made more fragmentary in the early stages of excavations. In the first instance, permission was granted only to attempt to trace the masonry of the Roman tower for a matter of three days. The semicircular front was located and the outline of the tower behind uncovered. This work was carried out by mechanical excavator as it seemed at that time that no further work was to be allowed. In this clearance the medieval layers in the area indicated on fig. 5 were partly destroyed. However, it was on the strength of what was uncovered in those three days that Trust Houses, greatly to their credit, agreed to the total excavation of the tower.

^{page 152 note 1} Hill, J. W. F., Medieval Lincoln (1965), pp. 127–8Google Scholar.

^{page 152 note 2} Ibid., p. 9.

^{page 152 note 3} The construction of the tower-cut oven is dated from the pottery found in the rubble return of its western side (fig. 10). This group is quite distinct from the pottery found in the filling of the oven, which belongs to a much later period (see pp. 168, 171).

^{page 156 note 1} Hill, J. W. F., Medieval Lincoln (1965), p. 121Google Scholar and fig. 9 on p. 122.

^{page 157 note 1} Antiquity, xlii (1968), 109Google Scholar.

^{page 157 note 2} Dr. J. P. C. Kent, Dept. of Coins and Medals, British Museum, identified the Claudian coins (nos. 3–6) and commented on their little-worn condition. The lettering and details of the figures are fresh and the coins have had little currency. Dr. Kent suggested a currency of about ten years as suiting their condition. Therefore the four coins may have reached the site at Lincoln by the early 50s, as the four coins are all issues of a.d. 41.

^{page 158 note 1} A melon bead of glass paste came from the filling of pit 2 (fig. 19, no. 25).

^{page 159 note 1} Hawkes, C. F. C. and Hull, M. R., Camulodunum (1947)Google Scholar.

^{page 160 note 1} For the two associated Claudian coins see p. 157, nos. 5 and 6.

^{page 163 note 1} Green clayey soil and sticky soils below mortar and rubble in centre of tower: No pottery is illustrated from these groups which, stratigraphically must belong to the period between the first and second stone structures. The samian ware and fragments of colour-coated ware box suggest a terminus post quern of a.d. 150–60. The mortar and rubble may represent the remains of the original floor of the third-century rebuilding of the gate. There are quite a number of nails from these layers.

^{page 163 note 2} See also p. 167, n. 1. The presence of ‘Dales ware type pottery in this group might suggest a slightly earlier date for the emergence of this distinctive type than originally suggested in Antiq. Journ. xxxi (1951), 159–60Google Scholar.

^{page 163 note 3} Gillam, J. P., Types of Roman Coarse Pottery Vessels Northern Britain, 3rd edn., 1970 (hereafter Gillam, 1970)Google Scholar.

^{page 164 note 1} For instance Gillam, 1970, nos. 70 and 71, dated a.d. 120–200 and a.d. 150–200 respectively.

^{page 164 note 2} Antiq. Journ. xxvii (1947), fig. 3, opp. p. 64. This volume (pp. 61–79) contains the report on the pottery kiln excavated at Swanpool 1½ miles south-west of Lincoln (hereafter referred to as Swanpool).

^{page 167 note 1} The term ‘Dales ware type’ (see also p. 163, n. 2) is used to describe vessels of which the rims spring straight from the shoulder rather than curving out from it, after Gillam, in Antiq. Journ., xxxi (1951), 154–64CrossRefGoogle Scholar. The rim itself is flat-topped or internally bevelled. The term ‘Dales ware type’ is used in preference to ‘Dales ware’, as this type of rim is now widely known from Lincolnshire and must have been made at a number of local potteries.

^{page 167 note 2} P. Corder, The Roman Pottery at Throlam, Holme-on Spalding Moor, East Yorkshire, 1930.

^{page 167 note 3} Antiq. Journ. xl (1960), 214–20Google Scholar.

^{page 167 note 4} Material from unpublished pottery kiln at Bourne, Lines., in Lincoln Museum; noted in ‘Archaeological Notes 1959/60’ published in the Lincolnshire Architectural and Archaeological Society Reports and Papers.

^{page 171 note 1} A sherd from this group which is not illustrated is worth individual comment. It is in a creamy fabric with small red inclusions and has a light-yellow glaze on the outer surface. It has been suggested by Mr. J. G. Hurst that this might be a French import of the thirteenth to fourteenth centuries.

^{page 171 note 2} The comments of Dr. G. C. Dunning on nos. 8 and 9 are gratefully acknowledged.

^{page 172 note 1} The writer is grateful to Mr. J. G. Hurst and Mr. S. Moorhouse, Inspectorate of Ancient Monuments, for comments on the pottery from the well. Mr. Moorhouse in a letter says that nos. 19, 22, 23, and 28 could be of sixteenth-century date. The Frechen jug, no. 24, with its all narrow neck and globular body is characteristic of the earlier Frechen types and is datable to the second half of the sixteenth century. The slip-ware sherd, if English, is not earlier than the early seventeenth century, and probably dates to the middle of the century. Nos. 26, 27, and 29 are typical of mid-seventeenth-century groups. The writer is also grateful to Mr. D. C. Mynard for drawing his attention to a mid-seventeenth-century group from Dover Castle published by him in Post-Med. Arch, iii (1969), 31–46Google Scholar.

^{page 172 note 2} The description ‘Tattershall type’ has been coined by Mr. J. G. Hurst to describe a group of pottery characterized particularly by the mottled green glaze on the outer surface. Sometimes a dark green glaze appears on the inner surface. This type was first recognized at Tattershall Castle, hence the name, but has since been found widely distributed in Lincolnshire. Its source is still unknown. It was also represented at The Park, Lincoln, 1968, in the upper filling of the Roman ditch.

^{page 182 note 1} Oswald, Adrian and Phillips, Howard, ‘A Restoration Glass Hoard from Gracechurch Street, London’, The Connoisseur (September 1949), pp. 30–6Google Scholar.

^{page 182 note 2} Loc. cit., p. 34. The fleur-de-lis border is echoed on the AT/RT glass dated 1578 in the Fitzwilliam Museum, Cambridge (Circle of Glass Collectors, Commemorative Exhibition 1937–1962, Victoria and Albert Museum, London (1962), Catalogue, no. 102, pl. 1); by the scrolled spearhead motifs of the top border of the KY glass dated 1583 (W. A. Thorpe, History of English and Irish Glass (London, 1929), vol. ii, pl. 1 b); and by the trefoil cresting of the ‘Barbara Potters’ glass dated 1602 (W. A. Thorpe, English Glass (London, 1961), pl. xvi b).

^{page 183 note 1} Cf. the glasses mentioned in n. 5 on p. 183. To these may be added the Dier glass of 1581 (Thorpe, History, pl. 1 a) and the GS glass dated 1586 (ibid., pl. 11 a).

^{page 183 note 2} Cf. the Dier glass (see n. 1), the Barbara Potters glass (p. 182, n. 2), etc.

^{page 183 note 3} Pit 209, site 11. Kindly shown to me by Dr. Colin Platt.

^{page 183 note 4} I am indebted to my daughter, Mrs. J. Stringer, for drawing this fragment to my attention, and to Mr. Norman Cook and Mr. Hugh Chapman for giving me the opportunity of examining it.

^{page 183 note 5} King Henry VIII's inventory of 1542 mentions occasional pieces which seem certainly to have been enamelled and gilt (Hartshorne, A., Old English Glasses (London, 1897), pp. 464–5Google Scholar), and examples have been excavated, e.g. in Southampton and Winchester: see also Charleston, R. J., ‘Pottery, Porcelain and Glass’ in (ed.) Edwards, R. and Ramsey, L. G. G., Connoisseur Period Guides, The Tudor Period (London, 1956), p. 84, pl. 46BGoogle Scholar.

^{page 183 note 6} King, W., ‘Eine Gruppe emaillierter venetianer Gläser im Britischen Museum’, Pantheon (1929), pp. 473–5Google Scholar.

^{page 184 note 1} Barrelet, J., La Verrerie en France (Paris, 1953), pp. 72–3Google Scholar, pls. xxxvi–xl.

^{page 184 note 2} Hetteš, Karel, ‘Venetian Trends in Bohemian Glass-making’, Journal of Glass Studies, v (1963), 38–53Google Scholar, particularly pp. 45–53, figs. 7, 12–24. fig.

^{page 184 note 3} Birdwood, G. C. M. and Foster, W., The First Letter Book of the East India Company, 1600–19 (London, 1893), p. 76Google Scholar.

^{page 184 note 4} See Daniels, J. Stuart, The Woodchester Glass House (Gloucester, 1950)Google Scholar, pl. vii, figs. 54–80.

^{page 185 note 5} R. J. Charleston, notes on the glass, in Huggins, P., ‘Excavations at Sewardstone Street, Waltham Abbey, Essex, 1966’, Post-Medieval Archaeology, iii (1969), 94–5Google Scholar, 35, particularly no. 2, and parallels cited there.

^{page 184 note 6} Cf. Stuart Daniels, op. cit., pl. 1, 1, 2, 14; pl. vi, 45, 47; pl. x, 3; pl. xi, 14. The Woodchester glasshouse probably flourished about 1590–1615.

^{page 185 note 1} Itinerarium Curiosum (2nd edn., 1776), p. 89 and pl. 54.

^{page 185 note 2} The original is in the Ashmolean Museum.

^{page 185 note 3} 1st edn. (1789), ii, pl. in, fig. 14; 2nd edn. (1806), ii, pl. vii, fig. 14.

^{page 185 note 4} Cf. the engraving by John Lodge in Gents. Mag. 1771, opp. p. 200. The original drawing was probably the inspiration (rather than vice-versa) for at least two paintings which, apart from minor additions of figures, 54. are virtually the same; one, by Sir Nathaniel Dance, is in the Usher Art Gallery, Lincoln, and the other passed through the London sale-rooms in 1937 (cf. Illustrated London News, 20th February 1937, p. 318, fig. 1).

^{page 185 note 5} The drawing is in the British Museum.

^{page 185 note 6} J.R.S. xxviii (1938), 182Google Scholar and pl. xxi.

^{page 185 note 7} Soc. Antiq. MS. 786, vol. 6, f. 22 (for the Willson collection, see p. 194, n. 4).

^{page 186 note 1} Arch. J. ciii (1946), 32Google Scholar and pl. via.

^{page 186 note 2} Ibid. 34 and fig. 6.

^{page 186 note 3} Mr. J. G. Hurst, F.S.A., of the Ministry was most helpful in making the necessary arrangements, while Mr. F. T. Baker, F.S.A., Secretary of the Lincoln Archaeological Research Committee, made the necessary contacts and enlisted local support with his characteristic skill and energy.

^{page 187 note 1} Thanks are due to the Corporation both for permitting the original excavation and for their subsequent sympathetic attitude towards preservation; Mr. A. Adlington, the City Engineer and Surveyor, was particularly helpful in arranging for the supply of men and equipment, and for allowing the excavation to encroach on the footpath to the east of the site.

^{page 187 note 2} Published by Sympson, E. Mansel in Lines. Notes and Queries, x (1909), pl. opp. 129Google Scholar; its significance as a medieval structure was noted by Richmond in Arch. Journ. ciii (1946), 32Google Scholar n.

^{page 189 note 1} It may be noted that the west gate of the Norman castle (pl. xxxvi b) has a front arch flanked by projecting walls to north and south—conceivably a later addition ?

^{page 191 note 1} A similar treatment can be observed on the massive blocks forming the north jamb of the east gate (pi. xix b).

^{page 193 note 1} R. E. M. and T. V. Wheeler, Verulamium, a Belgic and Two Roman Cities (1936), pls. xxii and xxiii.

^{page 193 note 2} Kindly contributed by Mr. F. T. Baker.

^{page 194 note 1} Gent. Mag. 1836 (i), 583–5Google Scholar.

^{page 194 note 2} Lincoln Museum no. 9295/06; it was published by Tuke as a chromolithograph, The Western Entrance in Roman Lindutn, 1836, and reproduced in Ian Richmond, Roman Britain (1947), col. pl. opp. p. 41.

^{page 194 note 3} Arch. Journ. ciii, 31–5.

^{page 194 note 4} MS. 786, vols. 1–20 being mainly notes and A–M mainly drawings. The collection was formed by Edward James Willson, F.S.A. (1787–1845), the Lincoln architect and antiquary, who was appointed County Surveyor for Lincoln Castle in 1833 and also advised the Dean and Chapter on the cathedral. For his life see Gent. Mag. 1855 (i), 321, and The Builder 6 Jan. 1855. The collection was bought by the Society in 1901 for £200 (Proc. Soc. Antiq. xix, 13).

^{page 195 note 1} This description is a conflation of Frederick Burton's account, with illustration, in Gent. Mag. (1836) (i), 583, the Tuke drawing, and unpublished drawings in the Willson collection (Soc. Antiq. MS. 786. 6, pp. 26, 27).

^{page 195 note 2} Op. cit., p. 26 (dated 6 p.m., Friday—presumably 15th April 1836, the day of the collapse).

^{page 196 note 1} F. H. T.'s italics.

^{page 196 note 2} Op. cit., p. 27; however, the title and scale on this measured drawing are apparently not in Willson's hand-writing.

^{page 196 note 3} Annales Lincolnenses (1850), i, 28. 9; this manuscript, formerly in Lord Monson's library at Burton Hall, is now in the Lincoln City Library.

^{page 196 note 4} Arch. Journ. ciii (1946), 34–5Google Scholar.

^{page 197 note 1} Op. cit., p. 384.

^{page 197 note 2} Op. cit., p. 26.

^{page 199 note 1} Mr. F. T. Baker has kindly contributed this paragraph.

^{page 200 note 1} Arch. Journ. ciii (1946), 32–3Google Scholar and fig. 7.

^{page 200 note 2} Collingwood, R. G. and Richmond, Ian, The Archaeology of Roman Britain (1969), p. 101Google Scholar.

^{page 200 note 3} Information kindly provided by Mr. Baker.

^{page 200 note 4} S. Wilkin (ed.), Sir Thomas Browne's Works, i 1836), 24.

^{page 200 note 5} J. B. Whitwell (Roman Lincolnshire (1970), p. 29) suggests there were not and Mr. Baker agrees.

^{page 200 note 6} Arch. Journ. cxvii (1960), 40–70.

^{page 200 note 7} ibid., figs. 1–3.

^{page 201 note 1} J.R.S. xxxix (1949), 57–78Google Scholar (west and north); J.R.S. xlvi (1956), 22–32Google Scholar (east); Arch. Journ. cxvii (1960), 40–70Google Scholar (south).

^{page 201 note 2} Arch. Journ. ciii (1946), 26–9Google Scholar; J.R.S. xxxix (1949), 57–78Google Scholar.

^{page 201 note 3} Ibid. 64–6 and figs. 6 and 8.

^{page 201 note 4} Arch, journ. ciii (1946), 29Google Scholar.

^{page 201 note 5} C.I.L. xiii, 6679.

^{page 201 note 6} J.R.S. xlvi (1956), 24–6Google Scholar.

^{page 201 note 7} Ibid. 26.

^{page 202 note 1} Arch. Journ. cxvii (1960), 50–4Google Scholar.

^{page 202 note 2} J.R.S. lv (1965), 205Google Scholar; lvi (1966), 203; lvii (1967), 181. A summary appears in Whitwell, J. B., Roman Lincolnshire (1970), pp. 26–8Google Scholar.

^{page 202 note 3} Britannia, i (1970), 284Google Scholar.

^{page 202 note 4} Arch. Journ. cxvii (1960), 50–1Google Scholar and figs. 4, 12–13 and 5, 30–3.

^{page 202 note 5} Ibid. 60, n. 2.

^{page 202 note 6} Ibid., fig. 4, 12.

^{page 202 note 7} Ibid. 55.

^{page 202 note 8} Y Cymmrodor, xli (1930), fig. 74, 227a–f.

^{page 202 note 9} Report forthcoming.

^{page 202 note 10} J. S. Wacher (ed.), The Civitas Capitals of Roman Britain (1966), p. 63.

^{page 203 note 1} J.R.S. xlvi (1956), 26Google Scholar; Arch. Journ. cxvii (1960), 51Google Scholar. See also Frere, S. S., Britannia (1967), p. 253Google Scholar, for a slightly variant view.

^{page 203 note 2} It appears in Claudian auxiliary forts, e.g. Valkenburg and Hod Hill.

^{page 203 note 3} J.R.S. xxxix (1949), 641–6Google Scholar and figs. 6 and 8.

^{page 204 note 1} Laur-Belart, R., ‘Vindonissa, Lager and Vicus’, Römisch-Germanische Forschungen, x (1935), pp. 24–8Google Scholar and Taf. 8.

^{page 204 note 2} A gate-plan discussed by Schonberger, H. in ‘Ein augusteisches Lager in Rödgen bei Bad Nauheim’, Saalburg Jahrbuch, xix (1961)Google Scholar. Cf. especially pp. 50–1 and Abb. 6, 7, 11.

^{page 204 note 3} Annals, xii, 32, referring to Camulodunum.

^{page 204 note 4} The strategic function is reflected in Cicero's description of a colonia as propugnaculum (pro Fonteio, 5, 13) and propugnaculum imperil (de lege agraria, ii, 73).

^{page 204 note 5} Cf. the classification of the gates at Rome (I. A. Richmond, The City Wall of Imperial Rome (1930), pp. 244 ff.).

^{page 205 note 1} Like the north gate at Caerwent. Cf. Arch. lix(19O5), 87.

^{page 205 note 2} de architectura i, v, 5.

^{page 205 note 3} Richmond, I. A., ‘Augustan Gates at Torino and Spello’, P.B.S.R. xii (1932), 52–62Google Scholar.

^{page 205 note 4} Schultze, R., ‘Die romischen Stadttore’, Banner Jahrbücher, cxviii (1909), 280–352Google Scholar, Taf. xiii.

^{page 205 note 5} Ibid., Tafs. xiii, xiv.

^{page 205 note 6} R. E. M., and Wheeler, T. V., Verulamium, a Belgic and Two Roman Cities (1936), pp. 63–75Google Scholar and pis. xxii, xxiii, lxxxix.

^{page 205 note 7} Frere, S. S., ‘Verulamium—Then and Now’ (Bulletin of the Institute of Archaeology, iv (1964), 61–82Google Scholar), 69–71.

^{page 205 note 8} Antiq. Journ. xli (1961), 65–7Google Scholar and fig. 1.

^{page 205 note 9} The most recent general account is G. Ulbert, ‘Das römische Regensburg’ in Gymnasium, Beiheft i (1960), 69 ff. (Germania Romana, i: Romerstddte in Deutschland.)

^{page 205 note 10} Schultze, op. cit., Taf. xvii.

^{page 205 note 11} Richmond, I. A., The City Wall of Imperial Rome (1930). pp. 120, 245Google Scholar.

^{page 205 note 12} For a recent and convenient review of the evidence, see Wightman, E. M., Roman Trier and the Treveri (1970), pp. 94–8Google Scholar and fig. 5. Ward-Perkins suggests a late third-or early fourth-century date, with a preference for the latter (Boethius, A. and Ward-Perkins, J. B., Etruscan and Roman Architecture (1970), pp. 520Google Scholar, 522). See also now Germania, li (1973), 178Google Scholar, n. 13.

^{page 206 note 1} Richmond, op. cit., pp. 245–6 and figs. 24 (Porta Appia, a first-class gate with double carriage-way) and 16 (Porta Nomentana, a second-class gate with single carriage-way).

^{page 206 note 2} Ibid., p. 254 and fig. 25.

^{page 206 note 3} Butler, R. M., ‘Late Roman Town Walls in Gaul’, Arch. Journ. cxvi (1959), 25–50Google Scholar, and especially pp. 41–2.

^{page 206 note 4} Bechert, Tilmann, ‘Römische Lagertore und ihre Bauinschriften’, Banner Jahrbücher, clxxi (1971), 201–87Google Scholar.

^{page 206 note 5} Ibid. 260 f.

^{page 206 note 6} Ibid. 239 f.

^{page 206 note 7} Ibid. 262–72.

^{page 206 note 8} Briefly discussed, ibid. 279 f.

^{page 206 note 9} Cf. Boethius and Ward-Perkins, op. cit, fig. 117 for a helpful reconstruction.

^{page 206 note 10} Arch. Aettana, 4th ser., xx (1942), 134–54Google Scholar.

^{page 207 note 1} Ibid., figs. 6–9.

^{page 207 note 2} Ibid. 145.

^{page 207 note 3} R. E. M. and T. V. Wheeler, op. cit., pl. lxxxix.