go 



THE PROBLEM OP LYNCttETS. 



certain that a terrace was first levelled and that then the channel 

 was hollowed out. It was not lined with puddled clay. Worked 

 flints were found in it, and small pieces of samian ware, together 

 with glass and pottery much less ancient. All these things 

 probably lay in the adjacent soil that slowly fell in and over- 

 whelmed the channel. 



Perhaps in the approaching summer a section may be made 

 much nearer Compton Vallance. Should the conduit be found 

 there also, a presumption will be raised that a stream once 

 issued from the hollows under the greensand rocks of that valley 

 comparable to the abundant flow that now escapes from a similar 

 formation in the Vale of Portesham. The channel in question, 

 which would be competent to convey a steady and constant 

 stream like that, would be wholly unable, in consequence of its 

 many windings, sharp curves, and shallow gradients, to carry 

 storm waters. 



By whom the channel was made, and what was its objective, 

 Poundbury or Dorchester, have yet to be determined. 



28. The last Diagram, M, shows the angles of slope of the 

 inner and outer walls of Poundbury near the great entrance. It 

 is hoped that by such observations it may be possible to 

 determine, in the absence of other evidence, the relative age of 

 prehistoric fortresses, of which the walls are built of known 

 materials, like the pastoral camps of Dorset. 



Jnyit^ s[vf >i_ 



yaUuArL. 



