More information : Not located. The cliff edge has been cut back by quarrying since Roman times. (1)
Between the known positions of milefortlet 23, established by air photo examination and geophysical survey, and the tower at NY 0267 3502, excavated by Bellhouse and identified by him as tower 26a, there is an accurately measured gap of 3474m. This area is occupied mainly by Maryport town and the industrial zone to the S of it; no trace of milefortlets or towers have ever been found here, and none are visible now. Bellhouse (authy 1) has suggested that the Roman fort at Maryport was in existence before or during the setting out of the frontier system, and that there was a 'dislocation' of the pattern of fortlets and towers at the fort. It is his belief that the series from the north terminated at the N corner turret of the fort, and then the measured sequence resumed from the S corner going S. There is no evidence that is not circumstantial to support this dislocation. The distance between the elements of the frontier need not be exactly one third of a Roman mile as is clear elsewhere along the Cumberland Coast and on Hadrian's Wall itself. Nevertheless, if the distance of 3474m is divided by 7, the result is 496.29m, or some 1.64m more than the standard one third of a Roman mile (494.65m). If the milefortlets/towers are plotted at 496m intervals in a straight line between MF23 and Bellhouse's T26a, tower 23a may be putatively located to NY 0420 3757 in a modern field. This grid reference cannot be proved at present, and cannot be treated as accurate. The frontier probably was not straight between MF23 and Bellhouse's excavated tower, and the distances between the elements need not be exact, though it is evident when these suggested sites are plotted on the OS 6-inch map that all are topographically feasible. In the absence of solid evidence for 'dislocation', and the fact that the pattern of towers S of the fort 'fits' the pattern to the N, it seems compelling that the sequence was not broken at the fort as suggested by Bellhouse. This means that T26a and T26b should be renumbered T25a and T25b. (2) |