Flanged bronze celt with sto[The maps of South-Eastern Britain and East Kent, like all maps of Ancient Britain, are inevitably inexact; but the errors are unimportant. The Dover cliffs, for instance, have lost by erosion, but one cannot say how much (see pages 528-30); nor is it possible to indicate the exact nature of the slight change which the coast has undergone between Sandown Castle and Walmer Castle (pages 521-5). Again, I have not attempted to delineate the coast west of Pevensey or west or north of Reculver precisely as it was in 55 B.C., because, even if such an attempt had been successful, nothing would have been gained for the purpose of this book. As far as possible, however, the maps represent the conclusions reached in the article on the configuration of the coast of Kent in the time of Caesar. The outline of Richborough harbour and of the estuary between Thanet and the mainland is intended to show approximately the high-water mark of spring tides. At low tide the channel was very narrow (page 519).]