By Mrs. M. E. Cunnington. 51 



neighboiirhood, and it belongs without doubt to the latter part of 

 the Late Celtic, or even to the early years of the Komano-British 

 period. There is evidence that at Knap, as elsewhere, this native 

 type of ware gave place in time to pottery characteristic of the 

 Eoman occupation, and within and about the plateau enclosure 

 pottery and other relics of this later period were found, proving 

 that the site continued to be occupied during the Romano-British 

 period, {tiee relic tables below.) 



The Mounds in the Plateau Enclosure. 



Within the area of the plateau enclosui-e there is a long artificial 

 bank or mound, that has sometimes been taken to be a long barrow. 



Adjoining this bank at the east end is a circular mound that has 

 also sometimes been mistaken for a barrow. 



For what purpose these mounds were raised is not clear, unless 

 indeed it was solely for protection. The situation is a very exposed 

 one, open to the north, and this bank would undoubtedly afford 

 considerable shelter, and no one knowing the situation could doubt 

 that any shelter from the winter winds as they blew across the 

 downs would be most grateful. Mr. Hadrian AUcroft mentions a 

 <ionsiderable vallum of earth, put up apparently only to serve as a 

 wind screen, along the western side of huts at Trewartha Marsh, 

 Launceston {Earthiuork of England, p. 206, note 3). 



The western end of the long mound rests on the silted-in ditch 

 of the old hill camp, so clearly this mound is of much later date 

 than the old camp. In a section cut across the neck of the bank 

 ■over the surface of the old ditch, nine fragments of pottery of the 

 bead rim bowl type were found, and a roundel of grey pottery ; in 

 addition to these were two small pieces of the thin red quality of 

 Koman pottery, and two pieces of thick pottery similar to that of 

 the large vessels found with Eoman pottery in the two fire holes. 



This shows that the bank must have been thrown up at some 

 time during or after the Itoman period, probably the former. 



On the surface it looks as if the mound at the east end had been 

 added to the original length of the long mound, but excavation 

 revealed no evidence of tliis having been actually the case. 



D 



