^{page 27 note 1} The drawings used to illustrate this article were made by my colleague, Mr. C. C. Pearce of University College, Reading, from pencil rubbings which I took from the original bowls in the Naples Museum while studying in Italy as Pelham Student of the University of Oxford. I wish to express here my thanks to him for the care and patience which he bestowed on a difficult and tedious task. My thanks are also due to the University of Oxford for further defraying, by a grant from the Craven Fund, the cost of preparing the drawings.

^{page 27 note 2} Not. Scav. 1881, pp. 300, f. and 322; 1882, p. 275, f.

^{page 28 note 1} The 37 bowls of the first group are numbered consecutively from 112892, 1/1882, to 112928, 37/1882, the remaining 53 from 112966 to 113018, 127/1882, the intervening numbers being no doubt assigned to the 37 lamps.

^{page 28 note 2} Ritterling, Hofheim, 1913, pp. 269 and 321, foot-note.

^{page 29 note 1} Cat. p. 46, M. 7–10.

^{page 29 note 2} Rottenburg, i, 7.

^{page 29 note 3} vol. i, p. 95, f.

^{page 29 note 4} Déchelette's no. 19 I failed to find.

^{page 29 note 5} Ritterling, Hofheim, 1913, p. 231 f.

It is convenient to give here the dates of the occupation of some sites frequently referred to below.

These dates are in most cases to be taken as only approximate. The end of the early period at Newstead is uncertain. There is some evidence that the site was occupied for some years after the recall of Agricola, but probably not beyond the very earliest years of the second century.

^{page 29 note 6} For its prevalence in the period 40–50 cf. Ritterling, Hofheim, 1913, p. 225 and Taff. xxvi and xxvii.

^{page 29 note 7} The left stroke of the penultimate letter is straight in the actual stamp.

^{page 30 note 1} cf. Knorr, Rottweil, 1907, Taf. xxvii, 9, where a bowl with the stamp of Reginus among the decoration has the stamp LVTAEVS.F on the smooth rim.

^{page 30 note 2} The same cursive stamp occurs on two moulds figured by Déchelette (i, p, 80, figs. 5, 9 and 60) and on a fragment of form 29 in the Museo delle Terme at Rome.

^{page 30 note 3} An explanation is perhaps needed of the use of this term in the following pages. The expression “décor à metopes” was first used by Déchelette, and describes accurately enough the upper frieze of a bowl of form 29 divided by vertical zigzag lines into small rectangular spaces containing alternately a group of figures and an ornamental design such as “arrowheads.” The meaning of the phrase is, however, extended to cover any decoration in which the field is divided into compartments by vertical lines, regardless of the fact that such an arrangement in the lower frieze quite ceases to bear any resemblance to the architectural feature from which the name is borrowed.

^{page 31 note 1} For these bowls compare Bushe-Fox in Archaeologia, lxiv, 295, fGoogle Scholar.

^{page 31 note 2} Déchelette, i, 86 f. 287, f.

^{page 31 note 3} cf. Ritterling, Hofheim, 1913, Taf. xxii, no. 216.

^{page 31 note 4} cf. C.I.L. xiii, part 3, i, 10010, 1374.

^{page 31 note 5} i, 287, f.

^{page 31 note 6} Rottweil, 1912, p. 3.

^{page 31 note 7} Wroxeter Report, 1912, p. 43. That this grouping is in general correct may be proved by a comparison of Knorr, Rottweil, 1912, Taf. i, ii, vii, ix and xi, with figs. 1, 2, 3, 4, and 5 below.

^{page 32 note 1} cf. Knorr, Rottweil, 1907, xiii, 2Google Scholar.

^{page 32 note 2} Ritterling, Hofheim, 1913, p. 249, f.

^{page 32 note 3} Déchelette, i, p. 303.

^{page 32 note 4} Bushe-Fox, p. 301.

^{page 32 note 5} Report, p. 242.

^{page 32 note 6} C.I.L. xiii, pt. 3, i, 100010, 2062.

^{page 33 note 1} Knorr, Das Kastell Ristissen, in Festschrift der K. Altertümer-Sammlung in Stuttgart, 1912, p. 57, where the date 60–83 or 85 is suggested.

^{page 33 note 2} The name occurs on Gaulish tombstones which suggest that it should be spelt Manduillus. On Samian it always appears abbreviated.

^{page 33 note 3} There are also two stamps in the British Museum, Cat. M. 812 on form 27, and M 942 on form 33, but neither of them can be attributed with certainty to this potter.

^{page 33 note 4} cf. no. 35 and no. 6.

^{page 33 note 5} Knorr, Rottweil, 1907, p. 65, Taf. i, 8.

^{page 33 note 6} Bushe-Fox, p. 300 and 308.

^{page 33 note 7} Festschrift, p. 57.

^{page 34 note 1} cf. nos. 9, 10, 11, 16 and 17.

^{page 34 note 2} i, 294.

^{page 34 note 3} Festschrift, p. 57.

^{page 34 note 4} C.I.L. xiii, pt. 3, i, 10010, 1378–9.

^{page 35 note 1} Jahrhuch für Altertumskunde, vol. vi (1912), p. 172 fGoogle Scholar.

^{page 35 note 2} Rottweil, 1912, p. 4.

^{page 35 note 3} cf. the bowl of Paullus (Knorr, Rottweil, 1907; Taf. xiv, 7, where the divided metope style is fully developed).

^{page 36 note 1} Knorr, Rottweil, 1912, Taf. i, 15.

^{page 36 note 2} cf. Hahnle, Die Relief-Kelche aus Haltern in the Mitt. d. Altertumskom. für Westfalen, vi, 1912Google Scholar.

^{page 36 note 3} cf. Forrer in Röm.-Germ. Korr. Blatt, May-June, 1912.

^{page 36 note 4} Knorr, _Rottweil, 1907, p. 27.

^{page 36 note 5} Rottweil, 1907, Taf. xiv, 6 and 12.

^{page 37 note 1} The drawing of no. 41 is inaccurate; the hinder animal should be the same dog as appears on no. 43.

^{page 37 note 2} C. Roach Smith, The Antiquities of Richborough, etc. pl. iii, 1.

^{page 38 note 1} Jahrhuch für Altertumskunde, vi, 1912, p. 178, Taf. iiiGoogle Scholar.

^{page 38 note 2} Knorr, Rottweil, 1907, xiv, 7Google Scholar.

^{page 38 note 3} Rottweil, 1912, xvi, 1, 14 and 15Google Scholar.

^{page 38 note 4} ibid. xix, 1, 2, and xx, 2.

^{page 38 note 5} Déchelette, i, p. 97, fig. 66.

^{page 38 note 6} Cat. p. 130, fig. 130, M. 480.

^{page 39 note 1} cf. Cat. nos. M. 484, M. 536 and M. 553.

^{page 40 note 1} vol. i, p. 98, fig. 67.

^{page 41 note 1} vol. i, p. 289.

^{page 41 note 2} Hofheim, 1913, p. 246.

^{page 41 note 3} cf. Ritterling, Hofheim, 1913, p. 240 and p. 246, where the statistics of the finds suggest a much earlier date for one of the potters.

^{page 41 note 4} Cat. nos. M. 96 and M. 831.

^{page 41 note 5} Knorr, Rottweil, 1907, p. 64.

^{page 41 note 6} Bushe-Fox, p. 301.

^{page 42 note 1} The first two numbers are taken from the bowls. The reference to Notizie degli Scavi is to the description of the several bowls, (A) to the first group (1881, p. 300, f.) and (B) (1882, p. 275, f.).

^{page 56 note 1} By an error this metope has been omitted in the drawing.

^{page 58 note 1} Misplaced in the drawing.

^{page 63 note 1} This bowl, owing to its decoration, was not available for a complete examination, and consequently the numbers affixed to it could not be ascertained.