192 ME. K. D. OLDHAM ON THE [May I9OI, 



these slopes would be slightly increased, and in a larger degree on 

 the south than on the north, so that the evidence of the surface- 

 slopes points to the river having flowed from north to south. 



A similar indication is given by the surface-features. As one 

 travels southward from the Dunmail Eaise the road descends into 

 the broad and open valley of Grasmere ; below Grasmere the valley 

 contracts somewhat, but is still that of a larger stream than the 

 one which formed the Dunmail Kaise, and, a little farther on, the 

 open valley of the Windermere Lake is entered. Travelling in the 

 reverse way, we descend from the pass into the Thirlmere Valley, 

 much narrower and steeper-sided than the valley on the other side 

 of the pass, and at the lower end of the lake this valley divides into 

 that of the Noddle Beck, through which the coach-road to Keswick 

 runs, and the St. John's Yale, through which the waters of Thirl- 

 mere find their present outlet. Both of these valleys are smaller 

 than the Thirlmere Valley, and St. John's Vale in particular has much 

 more the aspect of belonging to the headwaters of a stream large 

 enough to have carved out the gap of the Dunmail Eaise, than of 

 the lower part of its valley. It looks as if the two valleys represent 

 two tributaries, which united at what is now the lower end of 

 Thirlmere, to form a river, flowing southward to what are now 

 Grasmere and Windermere. 



It is noteworthy that the valley of the Noddle Beck is faced by 

 the depression between Saddleback and Skiddaw, now drained by 

 the Glenderaterra and Caldew, which may not improbably mark 

 the course of the upper waters of the Dunmail Kiver. It will 

 probably be impossible to establish this definitely, and it must 

 remain a suggestion ; but it is not imj)robable that the diversion of 

 the upper waters of the Dunmail Eiver into the Derwent, and their 

 removal to the west, may have caused such a decrease of power 

 as to make the old river no longer capable of coping with the 

 barrier, which was rising across its course lower down, and so caused 

 the division of its drainage-basin and the ultimate reversal of the 

 direction of flow of its upper portion. 



Against the original southerly flow^ of this river can only be set 

 the valley of the Wythburn, which joins the Thirlmere Valley, 

 leaving an acute angle between it and the Dunmail Valley. An 

 examination of the map shows, however, that the upper waters of 

 this stream flow in a valley running at right angles to the course 

 of the old Dunmail Eiver, while the lower portion alone bends 

 slightly northward, and the level at which this bend takes place 

 is below that of the crest of the Dunmail Eaise. It is, consequently, 

 possible that the northerly bend of the lower portion of the W^ythburn 

 Valley is of a later date than the separation of the old Dunmail Valley 

 into two portions, and due to the altered conditions of erosion set up by 

 that separation. In any case, the obliquity of this valley would not 

 preclude the possibility of a southerly flow of the river to which it 

 was tributary, and against it must be set the bifurcation of the valley 

 at what is now the lower end of Thirlmere, and the fact that on 

 the south side of the watershed all the larger tributaries join the 



.jII 



