^{page 192 note 1} D.A.J. lxi (1940), 26Google Scholar; lxxxii (1962), 21.

^{page 192 note 2} Antiq. Journ. xxxi (1951), 154Google Scholar.

^{page 193 note 1} This is true, for instance, of Derbyshire Ware and probably Dales Ware: below p. 202.

^{page 193 note 2} Nottingham University Museum.

^{page 193 note 3} Ibid.

^{page 193 note 4} Ibid.

^{page 193 note 5} Ibid.

^{page 193 note 6} Ibid.

^{page 193 note 7} Ibid.

^{page 195 note 1} Nottingham University Museum.

^{page 195 note 2} Newark Museum.

^{page 195 note 3} Analysis of trace minerals in the component clays suggests that in origin the clays are not dissimilar. The hardness and porosity of the oxidized and reduced fabrics are about equal. Details of the mineralogical and chemical analyses are to be included in a further study of east Midland fabrics.

^{page 195 note 4} The specimen at Ilkley (Proc. Leeds Phil. Lit. Soc. xii (1966), fig. 12, 86) does not seem to be in exactly the same fabric.

^{page 195 note 5} Swanpool, fig. 4, D 24–32.

^{page 195 note 6} Crambeck I, pl. iii, nos. 60–79.

^{page 197 note 1} On the date of the Swanpool kilns, see below, p. 200.

^{page 197 note 2} Very little late Roman pottery from either Lincoln or Ancaster has been published and this statement is based on examination of the collections at Lincoln and the University of Nottingham.

^{page 197 note 3} e.g. the common storage-jars and the large, wide-mouthed bowls.

^{page 198 note 1} Unpublished material in Oakham Museum,

^{page 198 note 2} Antiq. Journ. xvi (1936), 458, no. 38Google Scholar (if this is not an East Anglian vessel).

^{page 198 note 3} Jewry Wall, fig. 26, 21.

^{page 198 note 4} Gaps shown on the distribution map (fig. 3) are thus probably only apparent, especially in north Leics. and in south Lines.

^{page 198 note 5} Mainly straight-sided bowls: below, p. 200.

^{page 198 note 6} Unpublished material in Nottingham University Museum.

^{page 198 note 7} Material, including several wasters, in Newark Museum.

^{page 198 note 8} From excavations conducted by the University of Nottingham 1962–7. One waster is a typically third-century jar with lattice decoration; cf. fig. 2, 11. It is not certain that the known kilns north and south of the Roman town made the ware.

^{page 198 note 9} Unpublished material in Lincoln Museum.

^{page 200 note 1} Dated too early in the original report (Swanpool, pp. 78–9).

^{page 200 note 2} Crambeck II, 409.

^{page 200 note 3} Arch. Journ. cii (1945), 86Google Scholar, fig. 18, 16.

^{page 200 note 4} Antiq. Journ. xviii (1938), 370Google Scholar.

^{page 200 note 5} P.C.A.S., xlviii (1954), 38Google Scholar, no. 77.

^{page 200 note 6} Arch. Journ. cii (1945), 86Google Scholar, fig. 18, 16; cf. 20, 1.

^{page 200 note 7} Jewry Wall, fig. 19, 28–9.

^{page 200 note 8} University of Nottingham excavations in 1966.

^{page 201 note 1} Below, p. 205.

^{page 201 note 2} In general, P.S.J.S. xciv (1960–1), 126 ff J. P. Gillam, Carnuntina, 66, Map IV.

^{page 201 note 3} Cf. the products of one of the Lincoln kilns; P. Corder, A Romano-British Pottery Kiln on the Lincoln Racecourse (1950).

^{page 201 note 4} Some examples of these second- and early third-century vessels are illustrated in fig. 2, 7–11.

^{page 201 note 5} i.e. the painted mortaria and the Huntcliff cooking-pots.

^{page 202 note 1} Carnuntina, pp. 64 ff. Above, p. 192, nn. 1 and 2.

^{page 202 note 2} D.A.J. lxxxii (1962), 21Google Scholar.

^{page 202 note 3} D.A.J. lxxxii (1962), 41Google Scholar; lxxxvi (1966), 100. The current conception, outlined above, of a single Derbyshire Ware fabric is almost certainly too simple. From all over the north-east Midlands have come representatives of other over-fired fabrics not dissimilar in appearance and character. The whole question needs thorough exploration.

^{page 202 note 4} All three deposits are unpublished: from a late-second-century well at Dragonby, Lines, (information from the excavator, Mr. J. May); from deposits of a similar date at Ancaster (Univ. of Nottingham excavations 1967); and at Margidunum (excavations by the writer in 1966).

^{page 202 note 5} Fig. 2, 12–13.

^{page 204 note 1} e.g. Carrawburgh, A.A.4 xxix, 62 (fourth century); Rudchester, A.A.4 i, 110 (undated); Milecastle 50, CW. lii, 35 (third century).

^{page 204 note 2} Among other sites: Aldborough, , Y.A.J. clvii (1959), fig. 10, 3, 4, 19Google Scholar; Catterick, , Y.A.J. cliv (1957), 256Google Scholar, fig. 11, 32–5; Brough, Corder, and Romans, Excavations at the Roman Town of Brough, 1936, fig. 11,26–7.

^{page 204 note 3} Trans. Thoroton Soc. liii (1949), 1Google Scholar and pl. 1.

^{page 204 note 4} Univ. of Nottingham excavations, 1962–7.

^{page 204 note 5} Excavations by the writer, 1966–7.

^{page 204 note 6} Swanpool, fig. 5, H 1–19.

^{page 204 note 7} e.g. Bonner Jahrbucher, 163 (1963), 358, Abb. 11, 2, 3, 6, 7; E. Gose, Gefässtypen römischer Keramik im Rheinland, p. 541.

^{page 205 note 1} Crambeck I, pl. iv, 98–9. There is a slight difference in the rim-form.

^{page 205 note 2} Great Casterton III, fig. 18, 10.

^{page 205 note 3} The distribution evidence set out in the original report probably exaggerates the Torksey sphere of influence. Certainly the kilns are well placed for distribution along the Trent, navigable for Roman vessels well to the south-west of Newark, but, since the Torksey fabrics and vessel-forms are not exclusive to that site, the pattern of dispersal is not easily reconstructed. The distinctive and peculiar types, it might be noted, are rare and have been recorded locally only.

^{page 205 note 4} Oswald, A., The Roman Pottery Kilns at Little London, Torksey, Lines. (1937)Google Scholar.