and O·SEVERI, and early pottery. This stamp is no doubt that of the potter CRISPVS, which occurs at Novaesium, and also probably at Hofheim. First century.

35. CRISSI M. On bottom of a cup CRISSI O occurs at Zugmantel, also at Rheinzabern. Ludowici, Urnen-Gräber, p. 156.

36. CRVC[VRO]. (See page 213, Fig. 5.) On fragment of bowl, Type 37. Pit LXI. First century. The stamp occurs at Rottweil. Knorr (Taf. xi. 1 and 2) suggests that he is probably a potter of Banassac, of the time of Vespasian.

37. CVCCILL M. On fragment.

38. CVDI·M. On bottom of platter, Type 31. Baths.

39. DAGO. probably DAGOMARVS. On bottom of cup, Type 27. Pit LIV. For association see OF COTTO supra. First century. The stamp occurs on a cup of the same type found in London, Walters, Catalogue of Roman Pottery, M. 1631. DAGOMARVS occurs at Wiesbaden, at Ober-Florstadt, and also on dishes of Form 18 found in London. Probably Lezoux.

40. DIVICATVS. On bottom of cup, Type 27 or 33. Retentura. The stamp occurs at Bar Hill and at Corbridge, also at Silchester on a Pudding-pan Rock form (R. A. Smith, 'List of Potters' Marks'). Second century.

41. [D]IVIX. F. (See page 227, Fig. 3.) On exterior fragment of decorated bowl, Type 37. Potter of Lezoux. Second century. He must have been working as late as the Antonine period. DIVIX. F. occurs at Camelon [DIV]IX. F. at Birrens, also at Corbridge, South Shields, Silchester and London.

42. [DIVI]X. F. (See page 221, Fig. 3.) On exterior fragment of decorated bowl, Type 37. Inner ditches, West Annexe.