Bij Tom Burgess, Esq., F.S.A. 309 



purpose of a general survey of the country, though many of these 

 simple mounds or tumuli are found in low-lying situations where 

 they would have been of little use for such a purpose. The fortified 

 mound was different from these, though some of them had a slight 

 rampart round their base. The fortified mound had not only a 

 vallum and agger, of various extent, but had other earthworks 

 around them which show that they were intended for habitation, 

 for the residence of a number of people, and that means were pro- 

 vided for defence against marauders. There are other mounds 

 frequently of large size which were erected by the sides of rivers, 

 with a view first o£ indicating a ford, and subsequently to defend 

 the passage of the river against a hostile foe. There are also smaller 

 ones which were used, even in mediaeval times, as the residence of 

 the local St. Christopher, whose business it was, as shown in the 

 earliest dated woodcuts, to guide people across the river and to in- 

 dicate the ford at night by means of a lantern. It is somewhat 

 curious that whilst many of these mounds and the adjacent sites 

 were formed into Roman stations and mediaeval castles, they were 

 not always used, as at Devizes, for the substructure of a donjon or 

 keep. Two or three very familiar instances occur to me. At 

 Leicester the Eata Coritani of the Itineraries, the mound is situated 

 on the river side, within the Roman boundary, and also within the 

 Norman castle precincts, but there is no sign of there ever having 

 been a building upon it. At Warwick, the mound attributed to 

 Ethelfleda in the tenth century has the remains only of some four- 

 teenth century masonry on it. It is situated by the side of the Avon. 

 It marks a ford, but neither the De Newburghs, nor their Norman 

 descendants, seem to have erected any building — certainly not a keep 

 — upon it. At Canterbury the Norman keep is a long way from 

 the Dane John, and instances might be multiplied to a great extent, 

 but they tend to show that a mound, even if existing at the time, 

 was by no means a necessity of Norman fortification. I need only 

 mention Worcester, the city which at present is my home ; there 

 the mound was by the side of the river. It was wholly artificial, 

 and when it was removed a few years since Roman, Saxon, and 

 mediaeval relics were found, most of which have been figured ; but 



