^{page 135 note 1} Mr. Hawkes dated the Thundersbarrow pottery in bulk and wrote brief notes on it before the material reached the present writer, who wishes to acknowledge the use made of these.

^{page 135 note 2} The hand-made native ware was distributed through the lower 4½ ft. of the filling, but more occurred at the bottom than elsewhere.

^{page 136 note 1} Not the so-called ‘mica-dusted’ ware—the mica is mainly in the paste and can only be clearly distinguished with a lens.

^{page 136 note 2} Bushe-Fox, , Excavation of the Roman Fort at Richborough, 1st Report, 1926, p. 90Google Scholar.

^{page 136 note 3} ‘Roman pottery kilns at Sandford’, Archaeologia, lxxii (1921–1922), 225–42Google Scholar.

^{page 137 note 1} In subsequent references to N. F. pottery sites in this paper, the plate numbers are as they appear in the 1927 edition of Sumner's works (collected). The plate numbers in the separate papers published earlier are different.

^{page 140 note 1} Sumner, op. cit., 1927 edition.

^{page 142 note 1} G. R. Wolseley and R. A. Smith, Archaeologia, lxxvi, 1–40.

^{page 142 note 2} Eliot Curwen, Sussex Arch. Coll. lxxii, pp. 185–217.

^{page 142 note 3} Pottery similar to the Thundersbarrow native ware (class A) has been found at Porchester Castle during work of clearing and consolidation by H. M. Office of Works. This, together with comparable fourth-century material from Richborough, was exhibited at the Society of Antiquaries on 10th November 1932.

^{page 143 note 1} R. G. Collingwood, Archaeology of Roman Britain, p. 233, and figs. 67–70.

^{page 146 note 1} Langton, p. 83, and fig. 102, etc.

^{page 146 note 2} These grits are no longer chalk, presumably having been converted to lime during firing. This would indicate that the pottery was fired at a temperature over 550° C. The liberation of CO₂ during the conversion of the chalk to lime would account for the spongy texture of the fabric in places.

^{page 146 note 3} As explained by Dr. Curwen on p. 122, neither exact shape nor dimensions of these jars have been recorded with absolute certainty.

^{page 148 note 1} Sumner, op. cit., 1927 edition.

^{page 148 note 2} Lime with traces of phosphate.

^{page 150 note 1} See R. G. Collingwood, Archaeology of Roman Britain, p. 224.

^{page 150 note 2} E. Curwen, Sussex Arch. Coll. lxxii, p. 208.

^{page 150 note 3} Langton, p. 99.