Fig 7  Plan of Roman Fort at Reculver
(From  Arch. Journal lxxxvi, p 299)

as even then endangered, and Battely, who died in 1708, testifies that in his lifetime the waves were already devastating the north face of the fort. By 1784, when Boys made his plan, its north-west angle had wholly vanished, and very soon afterwards the rest of its northern side perished. Sea-walls have since saved much, but nearly a third of its area has been swallowed up. The rate of erosion has probably not been uniform, and we cannot now determine how far off to the north the shore lay in Roman times. But we may assume that the fort of Reculver stood a little way back from the open sea, on some sheltered inlet or harbour.
   The fort itself, when perfect, formed a nearly square enclosure with rounded angles, measuring internally about 570 ft. by 585 ft., and covering, exclusive of its walls, a trifle over 7½ acres.²¹  The walls still stand in places 8 ft. to 10 ft. high, and must originally have been  more than twice that height. They are in general 8 ft. or 8½ ft. thick, but broaden to 10 ft. at the bottom, owing to two offsets on the inner side. They consist of a facing — now mostly gone — of coursed Kentish ‘rag,’ in which the stones average 5 in. to 6 in. square, a core of flint, tufa, and sandstone rubble, some of it set slantwise in a common fashion, and thirdly a slight foundation, about 6 in. in depth,
   ²¹ These dimensions are taken from a note and plan contributed by Boys to Nichols’ Bib1. Topogr. Brit. i, 83—6, showing the site in 1781, when the north-east angle of the fort still survived. Boys’ original drawing is in the Soc. of Antiq. scrapbooks at Burlington House. The Ordnance Survey, followed by Fox, Arch. Journ. liii, plan, makes them less, 565 ft. by 565 ft. But the north wall must have vanished before the Ordnance Surveyors measured the site and their dimension must therefore be conjectural.