By Mavd E. Cunnington. 



431 



Samian, at a depth of 2|ft. from the surface. Under these cir- 

 cumstances it is not possible to attach any importance to the 

 presence of this piece of Koman pottery in the body of the rampart. 

 The ditch at this point was Q^ft. deep and 14ft. wide. For the 

 first 5ft. or 6ft. from the bottom it was filled up with a pure fine 

 chalk silt, and above that to the surface with an earthy mixed 

 silting in which there were many snail shells. At a depth of 3ft. 

 a piece of chert was found, and 6in. lower down an iron object 

 of unknown use. The piece of chert has a mammilla ted surface 

 and is perhaps from the Upper Green Sand. Its flat surface 

 is poHshed, and it appears to have been used as a rubber, while 

 the edges show signs of use as a hammer. 



Fig. 6. Iron object of unknown use, J. 



Section B.— 6ft. wide through Kampart only. 

 The rampart was here 5ft. 3in. high. As explained in reference 

 to Section A this cutting was made in order to see if it would 

 afford any evidence of the rampart having been built at two periods. 

 The section afforded no such evidence, its appearance being per- 

 fectly uniform throughout, with a well-defined old surface line. 

 A few fractured flints were found in the surface soil, but nothing 

 in the body of the rampart until the old turf line was reached. 

 The discovery in this of two ancient hearth sites alone repaid the 

 time and labour expended on the section. These two hearths 



