478 PROCEEDINGS OF THE GEOLOGICAL SOCIETY, [June 19, 



it had to go round, these south-east scratches being generally on 

 lower levels than those to the south-west. They would therefore 

 probably be made at the earlier or later part of the ice-sheet period — 

 most likely the latter ; for the ice at its maximum would obliterate 

 the effects of its earlier development. The south-west scratches seem 

 to lead on from those on the high ground north of the valley, and 

 may belong to the same set as those on Bowland Knotts, the change 

 in direction being accounted for by the form of the ground. On the 

 other hand, it is possible that they may be the result of a local glacier 

 later on, finding its way to the valley. 



These scratches running in directions transverse to one another 

 call to mind a suggestion which Professor Eamsay made to me 

 some two or three years ago, that it is possible that there may have 

 been undercurrents, so to speak, in the ice-sheet, caused by the form 

 of the ground. That this occurs on a small scale, at any rate in 

 glaciers, I am pretty well convinced. I saw an instance of it on 

 the Zardezan glacier at the head of the Val Pellina in the summer of 

 1869. A small glacier debouching from a higher level on to the 

 main ice-stream spread so far across it that the median moraine 

 resulting from the junction of the two swept round in a curve nearly 

 to the opposite bank of the main glacier, which was very far from 

 being a narrow one. There could be no doubt- that two bodies of ice 

 were moving in different directions, one above the other, for some 

 distance ; but I should feel inclined to hesitate before appljdng the 

 principle on so great a scale to explain the opposing scratches in so 

 broad a valley, especially as they may be explained by better-known 

 phenomena, as I have already shown. 



Leading on fi'om Bowland Knotts is another scratch, on the oppo- 

 site side of the H odder valley ; it trends across the valley and across 

 the ridge on which it lies. 



On the Watershed. — About seven miles west of Skipton lies the Great 

 Watershed, here on comparatively very low ground. It runs along a 

 range of hillocks rising to about 700 leet. They are composed chiefly 

 of drift; but in some parts the underlying limestone rises to the surface. 

 Here is the only scratched surface which I have yet found which shows 

 glaciation from west to east. This is not to be wondered at. It lies 

 between the basins of the Ribble and the Aire, at one of the lowest 

 passes in the Pennine range across the watershed of England. 

 Granted such a development of ice as I have already indicated, it is 

 quite impossible that some of the ice should not have been discharged 

 in this direction. I am aware of the difficulties that are raised about 

 the east side of the Pennine chain being so free from drift : but all 

 geologists agree that some, at any rate, does exist ; and the patches 

 that are left cannot be considered to have been dropped just as they 

 are, in isolation. The difficulty lies rather in the question, how has 

 the mass of it been removed ? That is a question which I cannot 

 undertake in this paper ; I do not know the ground sufficiently to 

 hazard an opinion ; nor, indeed, does the matter come within the 

 scope of this inquiry. 



Other scratches seem to run in the direction of East Yorkshire and 

 across the watershed, as about Gisburn, Gledstone, and Barnoldswick. 



