^{page 175 note 1} The work was arranged by the Ancient Monuments Inspectorate of the Ministry of Works; four men were employed, and help was also given by local volunteers, including boys from Caistor Gram- mar School. I should like to thank these for their help, and also the following: Mr. A. Holt, the owner of the property, who gave us every co-operation, and who has given the finds to Lincoln Museum; Mrs. E. Rudkin, of Willoughton, who gave her usual generous hospitality, and allowed me to draw on her invaluable knowledge of local archaeology; and Mr. F. T. Baker, Dr. Philip Corder, Mr. Sheppard Frere, and Mr. Graham Webster, who visited the site during the excavation, and contributed many ideas incorporated in this paper.

^{page 175 note 2} We are indebted to Dr. A. W. Woodland and Dr. V. Wilson of the Geological Survey and Museum for their advice on these matters, and for identification of stone samples submitted to them.

^{page 175 note 3} Still used locally, particularly the Eye Spring, in the corner of the churchyard, and the main Syfer Spring near by; there has in the past been some ‘spa’ attraction held out to visitors to Caistor, and a similar attraction in Roman times is not improbable.

^{page 177 note 1} ‘The Present State of Archaeology in Lincolnshire’, Arch. Journ. xci (1934), pp. 129–33Google Scholar.

^{page 177 note 1} ‘Roman Ancaster, Horncastle, and Caistor’, ibid, ciii (1946), p. 17.

^{page 177 note 3} ‘A Lost Inscription of Pre-Danish Age from Caistor’, ibid, ciii (1946), p. 95.

^{page 177 note 4} ‘The Re-organisation of the Defences of Romano-British Towns in the Fourth Century’, ibid, cxii (1956), pp. 38-40.

^{page 177 note 5} Itin. Curios., first edition 1724, pp. 96-97.

^{page 177 note 6} Ibid., posth. second edition 1776, pls. 19 and 20.

^{page 177 note 7} ‘Caistor Church and Town’, by Peter B. G. Binnall. See also Caistor R.D.C. official guide, where the Roman ‘camp’ is elongated to 1,000 ft. × 200 ft.

^{page 178 note 1} See p. 177, n. 2.

^{page 178 note 2} They were kindly examined by Mr. J. G. Hurst, M.A., F.S.A.

^{page 178 note 3} Hamilton-Thompson, A., ‘Pre-Conquest Church Towers in North Lincolnshire’, A.A.S.R. xxix, pt. 1, pp. 43 ff., 62-63Google Scholar.

^{page 180 note 1} Farther to the west beyond the wall (see below, p. 181) the configuration of the ground is now what it may have been before the wall was built (‘normal’ slope shown in fig. 1).

^{page 180 note 2} Layer M is apparently derived from this; it extends farther south than the footing is likely to have done, but was a disturbed layer, and may be mixed with the Roman or post-Roman levels south of the wall.

^{page 180 note 3} Shown on right of fig. 2 as the heavy dotted line ‘top of layer P elsewhere’.

^{page 180 note 4} The greater part of the stone used in all parts of the Caistor wall is Tealby Limestone, though there are also small amounts of Roachstone and Red Chalk, both of local origin.

^{page 180 note 5} Probably representing organic matter such as grass, etc.

^{page 182 note 1} I am indebted to Mr. Graham Webster, who examined the sherds, for this information.

^{page 182 note 2} In recent excavations under the church floor, near the junction of nave and south transept, for the installation of a heating system, chalk was found at 6 ft., with a layer containing Roman sherds at 4-4½ ft., sealed only by disturbed layers containing human bones.

^{page 180 note 3} Itin. Curios., first edition 1724, p. 96.

^{page 183 note 1} The upper dotted line infig.3; the lower dotted line shows the suggested line of wall.

^{page 183 note 2} If this theory is correct, then the area outside the Roman east wall, originally an open level space, has remained so in its function as Market Square. It is difficult to be sure of its intermediate history without further research, for this part of Caistor was largely destroyed by a disastrous fire in 1681, and largely rebuilt in the years following.

^{page 185 note 1} 1 ft. 8 in. long, 15 in. diameter, dressed flat at one end, the other end hollowed out in a parabolic shaped basin 9 in. deep, with 2-in.-wide rim, perhaps secondary, but not obviously so. It may be the tenon for a cap.

^{page 185 note 2} A recent trench, just behind the Syfer Spring, found bedrock at 5 ft., and the wall is probably farther back.

^{page 185 note 3} After the owner of it; the names of North, Cooper's, and Williams's Bastions given in this paper are only for convenience, and until such time as all the bastions can be numbered or lettered in sequence.

^{page 185 note 4} Scheduling by the Ministry prevented its complete destruction.

^{page 185 note 5} Neither of these must be confused with the two kinds ofmodern mortar applied for its consolidation.