CHAP. IV. OF ANCIENT BUILDINGS. 199 



filled up the interstices in the overlying 

 rubbish with their castings, which were after- 

 wards accumulated to a thickness of nearly 

 three inches over the whole surface. If we 

 add to this latter amount the mould between 

 the fragments of stones, some five or six 

 inches of mould must have been brought up 

 from beneath the concrete or tiles. The con- 

 crete or tiles will consequently have subsided 

 to nearly this amount. The bases of the 

 columns of the aisles are now buried beneath 

 mould and turf. It is not probable that 

 they can have been undermined by worms, 

 for their foundations would no doubt have 

 been laid at a considerable depth. If they 

 have not subsided, the stones of which the 

 columns were constructed must have been 

 removed from beneath the former level of 

 the floor. 



Chedworth, Gloucestershire. The remains 

 of a large Roman villa were discovered here 

 in 1866, on ground which had been covered 

 with wood from time immemorial. No 

 suspicion seems ever to have been enter- 

 tained that ancient buildings lay buried here, 

 until a gamekeeper, in digging for rabbits, 





