^{page 112 note 1} For accounts of the ‘villa’ and remains found in it see Luard, Archaeologia Cantiana, ii. 1–16, and Payne, Collectanea Cantiana, 179 ; for the inscribed tiles see Luard, p. 4, and plate 6. (In this plate the stamp appears three times, in no case perfect. At the top is a vestige of parietalem and a whole second line (banu. cabriab) ; the third line is wanting. In the middle, inversely to the top stamp, is part of the line -icavit, a whole line (cabriabanu. ca), and very faint vestiges of parietalem. At the bottom are traces of the lines icavit and cabriabanu. The tile has plainly been stamped three times, and care has not been taken to prevent the stamps from overlapping and obliterating each other's edges.) Luard succeeded, however, only in reading one line, Cabriabantu. From him it is quoted by Roach Smith, Roman London, 115, who cites wrongly from memory, and by Huebner, Corpus Inscr. Latin., vii, no. 1238.

^{page 112 note 2} 3rd series, iv. 247–56.

^{page 115 note 1} The list in Arch. Aeliana, iv. 251, requires to be corrected by the substitution of nos. 3 and 4 above for two entries of Coustantius.

^{page 116 note 1} For the Pan Rock see Mr. R. A. Smith's papers in the Proceedings, xxi. 268 and xxii. 395. The deposit of plain Samian found at the fort of Alteburg-Heftrich (not Pfünz) on the German Limes near Wiesbaden only contained ten Samian vessels and one other.

^{page 125 note 1} Bonner Jahrbücher, lxxxvi, plate iv, fig. 20, 170, 220.

^{page 125 note 2} Ludowici, Urnengrüber römischer Töpfer in Rheinzabern, 274 ; and his Stempelbilder römischer Töpfer, 252.

^{page 127 note 1} Déchelette, Vases ornés de la Gaule romaine, ii. 124, no. 798.

^{page 127 note 2} Walters, Cat. of Roman Pottery (Brit. Mus.), no. M1546. fig. 215 ; for the evidence of moulds see p. xxvi, note 2.

^{page 128 note 1} J. Collingwood Bruce, The Roman Wall (1851), 441.

^{page 129 note 1} Early Iron Age Guide (British Museum), 141, fig. 135.