^{page 151 note 1} The proofs of this report were read by Miss M. V. Taylor, and for this kindness, as for her pertinent criticisms, the writer would record his sincere thanks.

^{page 151 note 2} Speed, The History and Antiquity of Southampton with Some Conjectures concerning the Roman Clausentum, written in manuscript about 1770, and edited by Miss E. R. Aubrey, M.A., as a publication of the Southampton Record Society, 1900. It may be mentioned that, though Speed's identification is almost certainly correct, his topographic evidence is faulty, i.e. he identifies Regnum (Chichester) with Ringwood, Hants.

^{page 151 note 3} Warner, Attempt to Ascertain the Situation of the Ancient Clausentum (1792).

^{page 151 note 4} Englefield, A Walk through Southampton, 2nd edition, 1805.

^{page 153 note 1} A number of finds was recovered from the old surface beneath the outer bank during the partial destruction of the earthwork in 1923, including an iron pick, a bronze ring, pottery, and five coins. The coins were: two of Gallienus, c. A.D. 266 (in. DIANAE CONS AVC, stag r.; APOLLINI CONS AVC, griffin 1.) and one each of Tetricus, c. A.D. 270–3 (in. SPES …….); Gratian, c. A.D. 368 (in. ? GLORIA NOVI SAECVLI); probably fourth century A.D. See Proc. Hants Field Club, ix, i, 391. Compare the dating evidence of these coins with that adduced for the inner masonry wall (p. 161 of present report).

^{page 153 note 2} If this description is compared with that of the wall revealed during the present excavations, it would seem that the structure was of a similar build throughout its length.

^{page 153 note 3} Englefield's description is: ‘… at the distance of about 78 feet from the northern tower, another semi-circular tower, or butress, was discovered, of 24 feet in diameter’. Englefield shows this tower on his plan just south of the Southampton—Portsmouth road, and confirms its position in his text. If this is correct, Englefield's dimension of 78 ft. is wrong, since on his plan, according to the scale he gives, the distance from the intermediate tower in question to the northern angle is about 400 ft.

^{page 155 note 1} In 1939 excavations for air-raid shelters in the angle south of the junction of Hawkeswood Road and Bitterne Road revealed Roman occupation layers; this does suggest that the Town Wall should be located farther to the south-west, as shown on the site plan.

^{page 155 note 2} P.S.A. xxxi, 1918–19, 37–9.

^{page 155 note 3} J. S. Davies, History of Southampton (1883).

^{page 155 note 4} To facilitate reference to the sections, a number, preceded by the letter L, is appended to the written description of each stratified Roman level, The corresponding level on the sections is indicated by a similar number enclosed within a circle.

^{page 157 note 1} For the latest study of the New Forest Potting industry see C. F. C. Hawkes, ‘An unusual find in the New Forest Potteries at Linwood, Hants’, Antiq. Journ. xviii, 113.

^{page 157 note 2} It may be mentioned that, here and there on the surface of the mortar beds, where the flints had been removed, there could be seen, preserved in the matrix, the imprints of the builders' nailed footwear.

^{page 165 note 1} For references to the then unpublished Colchester material, and for much other help with this section, the writer is indebted to Prof. Hawkes.

^{page 165 note 2} Sussex Arch. Coll. lxxvi, 138–45.

^{page 166 note 1} Antiq. Journ. xviii, 262–77.

^{page 166 note 2} Casterley Camp (Devizes Mus. Cat., pl. xxvii, 6–8); Oare (ibid., pl. xli b); and Highfield, Salis-bury (Wilts. Arch. Mag. xlvi, 605–6).

^{page 166 note 3} Worthy Down, Twyford Down, Bury Hill (Period 3), Oare, Highfield, and Sud Moor (Isle of Wight) all cover the field.

^{page 171 note 1} Mr. M. R. Hull, who has kindly examined the sherds, writes: ‘It seems to me that you have perhaps a local fabric, which will have to be dated on its own merits and on those of what is found with it. The date might be third or fourth century; the great difficulty being to say when the wares known to us as prevalent in the 4th century first came into use.’

^{page 171 note 2} e.g. New Forest (Sumner, op. cit. 22 f.); Oxfordshire kilns, Sandford (Archaeologia, lxxii); Dorchester (Oxoniensia, i); Cowley (ibid. vi).