we must therefore assume a road to Richborough. And, fortunately, there is an obvious possibility. Two Roman roads, one running due north from Dover, the other due east from Canterbury, meet about a mile south—west of Richborough, at the hamlet of Each End. These roads can now be traced no farther. But Each End has yielded no Roman remains, and the two roads cannot have stopped there. They must have gone on a mile to Richborough. It is true that some 600 yards of the intervening mile is marshland, and may have been sea in Roman times. But it was more probably salt marsh, and it may well have been crossed by a causeway such as is common in similar situations elsewhere.
   From the structural remains of Richborough we pass to the smaller objects discovered on the site. These, though very plentiful, need not detain us long, for a detailed description of select examples forms a notable feature of the recent Excavation Reports. But attention may here be directed to a few objects which have a special intrinsic interest. (1) In the first place, mention has already been made of two groups of fragmentary inscriptions or remains of marble casing found chiefly in the vicinity of the platform (above, p. 27). The minute relics of the main inscription include the letters VM, a P, B or R, an M, PE, SV, and a portion of two lines with an A in the upper and M . M or M . N in the lower.⁴²  Nothing can at present be made of these. The secondary group consists mostly of small numeral letters which have also been cited (Pl. IV). To these we may add the fragment found by Professor Garstang in 1900 and bearing in letters one inch high the end of a word . . . AVIT, which suggests the words [opus consumm]avit or the 1ike.⁴³  (2) The excavations of 1900 also yielded from a trench just east of the ‘cross‘ an ingot of silver stamped with the moneyer’s name. In shape it is somewhat of an hour-glass outline, 4¾ in. long, 3¼ in. wide at the ends, and 2 in. at the centre ; it is thickest at the middle (¼ in.) and thins out at the ends (to about ⅛ in.) ; it weighs 10 oz. 4 dwt., or 317.8 grammes, about a Roman pound.⁴⁴  The metal has not been analysed, but, if one may judge from such tests as specific gravity, it is principally silver with a little lead or tin. In the middle it bears an inscription, in letters ²/₁₆ in. or ³/₁₆ in. tall :

EX   OFFI

ISATIS

Ex officina Isatis : ‘From the workshop of Isas (or Isaac).’ It belongs to a small group of similar objects, all agreeing in weight and shape and character. One instance has been found in London, others near Coleraine in north Ireland, and others again at Dierstorf, near Minden, in north-west Germany. The London and Coleraine ingots are dated by coins found with them to the end of the fourth or the beginning of the fifth century. The Dierstorf specimens may belong to A.D. 425-37. Silver at this period seems to have been given and taken in trade principally by weight. Even the regular silver coins of the time vary so much that they must have been more often appraised by the scales than by any nominal value. These one-pound ingots seem to
    ⁴²  Second Richborough Rep . 12.
     ⁴³  Arch. Cant. xxiv, 272 ; Eph. Epig. ix, 990 ; illustrated in the Second Richborough Rep. pl. xiii.
     ⁴⁴  Arch. Cant. xxiv, 272 (brief mention); Eph. Epig. ix, 1257, p. 640. The ingot is now in the Canterbury Museum. For similar ingots, see Corpus Inscr. Lat. vii, 1196, 1198; Brit. Mus. Guide to Roman Britain, 72 ; Willers, Numismatische Zeitschrift xxx, 211, xxxi, 367, and Bronzeeimer von Hemmoor (Hannover 1901), p. 221.