[995] Norman Conquest, iii. 182.
[996] Histoire du Château d'Arques, by A. Deville, pp. x., 412 (Rouen).
[997] Ed. Howlett, p. 106.
[998] Cours d'antiquités monumentales (1835), v. 227, 228.
[999] Colchester, in Archæologia, to which he refers, was attributed to Edward the Elder, and Rochester was, of course, as yet, believed to be the work of Gundulf.
[1000] Compare Professor Freeman on Falaise: "More probably, I think, of the twelfth than of the eleventh [century]" (Norm. Conq., ii. 175).
[1001] Château d'Arques, pp. 307-312.
[1002] Ibid., pp. 48, 267.
[1003] Compare the "castrum in cacumine ipsius montis condidit" at Arques with the "castellum novum super flumen Tyne condidit" at Newcastle.
[1004] Compare, on this point, the acute criticism of Dr. Bruce (repeated by Mr. Freeman) that "Wace (v. 12,628) speaks of the horse of William Fitz Osbern [in 1066] as 'all covered with iron,' whereas in the [Bayeux] Tapestry 'not a single horse is equipped in steel armour; and if we refer to the authors who lived at that period, we shall find that not one of them mentions any defensive covering for the horse.'" Compare also the expression of William of Malmesbury, who lived and wrote under the tower-building king, that the Norman barons took advantage of the Conqueror's minority "turres agere," these being the structures with the building of which the writer was most familiar.