Yol. 6$.^\ OEIGIN OF CERTAIN" CANON-LIKE VALLEYS. 481 



The occurrence of isolated mounds, often capped by gravel, in 

 the area of the suggested lakes, is a feature common to them all. 

 Mr. Pocock mentions such mounds, standing up like islands in the 

 Cheshire plain, 1 and they are common near Oxford. While in some 

 places inequalities would be levelled by the accumulation of sediment, 

 in others, as the overflow-channels were deepened and the lakes 

 became shallower, what I venture to call ' lacustrine denudation ' 

 might, I think, have taken place. 



The origin of these mounds is not easy to explain on the hypo- 

 thesis of fluviatile erosion ; but no such difficulty presents itself if 

 we regard them as portions of the bottom of an ancient lake, the 

 level of which has been gradually lowered. 



IY. Harrington Dingle. 



There is a canon near Chirbury in Shropshire, known to Palaeozoic 

 geologists as Harrington Dingle, which seems to have originated as 

 an overflow-channel to a small Glacial lake. Running parallel to, 

 and along the edge of, a depression in the Wenlock Shales, sub- 

 sidiary to the valley of the Upper Severn, it forms a narrow, straight, 

 and sharply- cut gorge, the sides of which are so steep that on our 

 visit to it Mr. Rastall and I had some difficulty in scrambling down 

 to the bottom. Fig. 4 (p. 482) gives a contoured plan of the 

 immediate neighbourhood. 



Two small streams, Camlad and Caebitra, draining the double- 

 headed commencement of the valley in question, meet at Church 

 Stoke : the original course of the drainage probably followed the 

 Wenlock Beds northward towards Chirbury, being subsequently 

 diverted, and compelled to invade the Ordovician country on the 

 east. 



The triangular depression, south of Church Stoke, an area now 

 without natural outlet (see fig. 4, p. 482), seems to have been the 

 site of a Glacial lake, indicated at Lydham by distinctly-bedded 

 brickearth containing fair-sized boulders of an Arenig type, which 

 suggest the action of floating ice. Glacial Drift, with erratics of 

 Palaeozoic grit and volcanic ash, occurs around Montgomery and 

 Chirbury, near which localities the country has a hummocky and 

 morainic appearance. 



It would seem, therefore, that the lower ground north of Church 

 Stoke being blocked by the ice-stream of the Severn Yailey, the 

 drainage of the region south of that place was forced against the 

 higher Ordovician land, probably flowing at first between it and 

 the ice, but eventually excavating a channel through the former, in 

 places 100 feet deep and cut down to base-level, which has per- 

 manently captured the Camlad stream. 



Many such overflow- channels may no doubt be found elsewhere 

 in the glaciated area, and it is not necessary to discuss them all. 



1 ' The Geology of the Country around Macclesfield, &c.' Mem. Geol. Surv. 

 1906, p. 95. 



2. Ai 2 



