[327] Mr Clark thought there was another motte in the earthworks outside the walls, though he expresses himself doubtfully: "I rather think they [the mounds outside the city ditch] or one of them, looked rather like a moated mound, but I could not feel sure of it." _Arch. Cantiana_, xv., 344. Gostling (_A Walk about Canterbury_, 1825) says there were _two_, which is perhaps explained by a passage in Brayley's _Kent_ (1808), in which he describes the external fortification as "a lesser mount, now divided into two parts, with a ditch and embankment." P. 893. Stukeley's description (circa 1700) is as follows: "Within the walls is a very high mount, called Dungeon Hill; a ditch and high bank enclose the area before it; it seems to have been part of the old castle. Opposite to it without the walls is a hill, seeming to have been raised by the Danes when they besieged the city. The top of the Dungeon Hill is equal to the top of the castle." _Itin. Curiosum_, i., 122. It is of course not impossible that there may have been two mottes to this castle, as at Lewes and Lincoln, but such instances are rare, and it seems more likely that a portion of the bailey bank which happened to be in better preservation and consequently higher was mistaken for another mount. Mr Clark committed this very error at Tadcaster, and the other writers we have quoted were quite untrained as observers of earthen castles. At any rate there can be no doubt that the Dane John is the original chief citadel of this castle, as the statements of Somner, Stukeley, and we may add, Leland, are explicit. The most ancient maps of Canterbury, Hoefnagel's (1570), Smith's (_Description of England_, 1588), and Grose's (1785), all show the Dungeon Hill within the walls, but take no notice of the outwork outside. [328] _Archæologia Cantiana_, xxxiii., 152. [329] _Ibid._, xxi. [330] _Close Rolls_, i., 234b, ii., 7b, 89. [331] Now, to the disgrace of the city of Canterbury, converted into gasworks. [332] For instance, at Middleham, Rochester, Rhuddlan, and Morpeth. [333] _Beauties of England and Wales, Kent_, p. 893. [334] The passages from the _Pipe Roll_ bearing on this subject (which have not been noticed by any previous historian of Canterbury) are as follows:-- 1166-7. In operatione civitatis Cantuar. claudendæ £5 19 6 " Ad claudendam civitatem Cantuar. 20 0 0 1167-8. Pro claudenda civitate Cantuar. 5 1 1 1168-9. In terris datis Adelizæ filie Simonis 15 solidos de tribus annis pro escambio terræ suæ quæ est in Castello de Cantuar. 0 15 0 1172-3. In operatione turris ejusdem civitatis 10 0 0 " In operatione predicte turris 53 6 8 " Summa denariorum quos vicecomes misit in operatione turris 73 1 4