ROMANO-BRITISH NORFOLK APPENDIX : THE ROMAN BANK The embankments round the head of the Wash, which have been attributed to the Romans, lie principally within the counties of Cambridge and Lincoln. Of the whole reputed length of 150 miles, little more than one-tenth falls to Norfolk. This is the 'Roman Bank ' which runs round the Hundred of Freebridge Marshland from the Ouse at Lynn to the Nene at Wisbech. Its appellation, so far as I can make out, is not very old, and probably dates from some antiquary ; but it is itself a genuinely ancient work. Unfortunately direct evidence of its age is wanting. No Roman remains have been recorded from Marsh- land, except a suspicious aqueduct, etc., at Walpole (see above) ; but one or two coins have been found at Walsoken close to Wisbech, and at Wisbech itself there are traces of per- manent Romano-British occupation. This, however, would not prove the ' Roman Bank ' to be Roman, and the evidence definitely relating to this and similar embankments is not at all satisfactory. For example, the late Professor Babington, in his Ancient Cambridgeshire (ed. 2, p. 89), refers to a Roman vase as found ' in the Roman Sea Bank,' near Tydd St. Mary and Tydd Gout, in Lincolnshire, but close to Wisbech. This would, of course, be evidence that at least this sea bank might plausibly be considered Roman. But Professor Babington added that the urn was in Wisbech Museum, and when I went there and asked for the urn, it proved to be beyond question medieval. In default of direct evidence, we are therefore forced back, at least for Norfolk, on the general and a priori point of view. If the Romans did not build it (people ask), who did so, or who could have done so ? That, however, is not a satisfactory position. We know far too little about the early history of the Fens in detail, and we can hardly venture to assert that at no period between the fifth and the eleventh centuries could sea walls have been built. On the other hand, if the Romans really erected a con- tinuous embankment of 1 50 miles in length, it is probable that we should possess some definite memorial of it in the shape of an inscription or some reference to it in a historian. For the present, and in default of real evidence, the general question must remain unsettled, while, as far as Norfolk is concerned, we must admit that the actual remains recorded do not prove the ' Roman Bank ' to be certainly, or even probably, a Roman work. See E. M. Beloe in Norfolk Archaology, xii. 311 ; Journal of the British Archaological Association, xxxv. 12, 80, 177, 352. 323
