194 MR. L. J. WILLS OjS' LATE GLACIAL AXD [June I912, 



and now runs in a gorge in Triassic rocks at both places. Thus it 

 affords another typical instance of a superimposed drainage-system : 

 for the present course is determined by its relation, not to the 

 rocks through which it flows, but to the deposit of Drift super- 

 imposed upon them and since eroded away. The valleys of the 

 Clywedog and Alyn in the neighbourhood of Wrexham show that 

 even the courses of the tributaries fall under the same system. 



It is interesting to notice that, by the diversion at Chester, the 

 Dee to within a few miles of its mouth maintains its reputation for 

 escaping from its pre-Glacial course. 



(8) The Relationship of the Two Buried Valleys. 



With regard to the relationship between the buried valley near 

 Chester and Bangor and that which has been traced through Chirk, 

 if we assume that 30 feet below sea-level at Eodens Hall is the 

 greatest depth of the pre-Glacial valley thereabouts, it is clear that 

 the river at that period dropped about 150 feet between there and 

 Queensferry,^ whereas at the present day the fall is only about 20 

 or 25 feet. This would seem to indicate that the valley was a 

 5'Oung one and not yet graded down to the extent that the present 

 Dee is. 



Further, it appears that, since the river is now flowing near 

 Cefn at approximately the pre-Glacial level, the result of the Chirk 

 diversion is a lessening of the fall between there and Eodeus Hall 

 of about 50 feet. This may represent the short-circuiting of a 

 long bend of the river, such as may well have existed when the 

 Dee flowed south through Chirk and 8t, Martin's ILoor. At the 

 same time one cannot be sure that this is the case, because the 

 pre-Glacial 'thalweg' below the Cheshire plains is so much more 

 abrupt than the present one, that the fall from Cefn to Eodens Hall 

 may have been correspondingly steep. It is worth noticing in this 

 connexion that, if an uplift took place, it is in the softer rocks of 

 the Cheshire plains near the mouth of the Dee that the speedy 

 excavation would occur, whereas in the harder Silurian and Lower 

 Carboniferous rocks of the upper valley little effect might be 

 evident. This hypothesis agrees with the foregoing observations if 

 it be assumed that the river which cut out the Yale of Llangollen 

 reached the sea near Chester. 



The balance of probabilities is in favour of this having been the 

 case, yet absolute proof is not at present forthcoming. We have 

 traced the pre-Glacial valley southwards through Chirk into what 

 is now the drainage-area of the Severn, and from the sea to Eodens 

 Hall. Between these tw^o points the country is deep in Glacial 

 Drift, and we have to rely on wells and borings for our evidence as 

 to its thickness. I know of no well in the Ellesmere district that 

 touches solid rock. Only a poor approximation, therefore, to the 



1 The following arguments assume that the pre-Glacial valley was the 

 product of river-erosion. They will not bold if Glacial overdeepening has 

 .played a prominent part in forming the buried valley. 



