^{page 2 note 1} Mr. T. Davies Pryce kindly supplies the following commentary on the two sherds:

(i) Form 37. The glaze is fairly good, but worn. The two compartments are divided by a wavy line ending in a spiral bud or spike. The wavy line is not uncommon in the period Trajan-Hadrian, but is relatively infrequent in Hadrian-Antonine times. Figure to r. with ‘winged’ boots; almost certainly the Warrior (Déchlette 117, Lezoux; compare also D. 110, Libertus). Horizontally arranged, centrally constricted leaf-ornament. I cannot find an exact parallel, but this type of ornament is found from Flavian to Antonine times (cf. Oswald and Pryce, Terra Sigillata, vi, 11, OF COTOI Wroxeter; Knorr, , Töpfer und Fabriken … des ersten Jahrhunderts, 1919, 27,Google Scholar OF COTOI; Knorr, , Rottweil, 1912, xix, i, 2;Google ScholarBrit. Mus. Cat. of Rom. Pottery, 1031, Doeccus of Lezoux). The Caerleon leaf is, I think, more nearly related to the South Gaulish types than to the Antonine ones. On this account I should hesitate to place the fragment later than, say, about A. D. 120. The Libertus type of Warrior is also consistent with this dating.

(ii) Form 31. I have the greatest difficulty in diagnosing the date of these plate-fragments. The heavy foot-stand occurs both early (cf. Wheeler, , The Roman Fort near Brecon, S. 6,Google Scholar OF PRIMI) and late; so also the rouletting of the basal interior. The plate had evidently no very pronounced internal convexity of the base, as in the dish-like type of 31 (Ludowici Sb; Oswald and Pryce, xlvii, 3). On the other hand the rouletting is of the coarse second-century class. For these reasons I am inclined to place the plate mid-way between the well-known large plates with rouletted basal interiors, of the first century, and the later examples of form 31 with rouletted interiors, of the second century.

Let us say ‘Trajan-Hadrian’ for both pieces.

^{page 3 note 1} Essex Arch. Soc. Trans., N.S., iii, 273. I am greatly indebted to Mr. M. R. Hull, Curator of the Colchester and Essex Museum at Colchester, for preparing the new drawings of the coffin and the vessels which it contained—the latter here illustrated for the first time. The neck of the glass phial is scarcely datable, but the pot, of grey ware with a heavy, overhanging rim, is a characteristic Colchester type which begins shortly after the middle of the first century and lasts on until the latter part of the Antonine period. The later examples are somewhat slimmer than the earlier ones; the present example should not be very much later than c. A. D. 150 and suggests an unusually early date for the burial, interment by inhumation being rare before A. D. 180–200.

^{page 4 note 1} A. Allmer, Découverte de monuments funéraires au quarlier de Trion; Steyerd, A., Histoire de Lyon, i, 351;Google ScholarEspérandieu, E., Bas-reliefs … de la Gaule romaine, iii, p. 40, no. 1797.Google Scholar More than thirty of these cones are preserved in the museum at Lyon.

^{page 4 note 2} Linckenheld, E., Les stèles funéraires en forme de maison chez. les Médiomatriques et en Gaule (Publications de la Faculté des Lettres de l'Univ. de Strasbourg, fasc. 38. Paris, Soc. d'Édition: Les Belles-Lettres; Oxford, Univ. Press, 1927), 129,Google Scholar citing Coutil, , Mém. Soc. Préhist. Française, iv, 1919;Google Scholar and ‘Notes archéologiques’ in Bull. Soc. Antiq. Quest, 1917 and 1919.Google Scholar I am indebted to Miss M. V. Taylor for this reference.

^{page 4 note 3} Atti della R. Accademia dei Lincei: Notizie degli Scavi di Antichità, 1913, 273.Google Scholar

^{page 4 note 4} Ib., 1921, 191; hence Ashby, T., Times Literary Supplement, Dec. 21, 1922, 858,Google Scholar col. i. I am indebted to Professor R. C. Bosanquet for this reference. Incidentally, the occurrence of a cremation-burial at the end of the third century is noteworthy.

^{page 5 note 1} iv, 10.

^{page 4 note 2} Frazer, J. G., Pausanias's Description of Greece, v, 227.Google Scholar

^{page 7 note 1} Evans, A. J., Archaeologia, lii, 326.Google Scholar