486 PROCEEDINGS OP THE GEOLOGICAL SOCIETY. [June 19, 



This direction is rudely represented by the glacial scratches on 

 the rocks beneath. In the northern quarter of the map this boundary 

 may be carried along the Gatebeck Brook, as my friend Mr. Hughes 

 tells me. West of this you find Shap Granite and other Lake-dis- 

 trict rocks, but none to the east. The boundary runs thence towards 

 Lancaster. I have found such boulders in Till down the coast by 

 Hest-bank and Bolton-le-Sands. South of Lancaster it probably 

 takes the direction shown by the scratches until past the Central 

 Fells. Here it seems to turn a little more to the east ; for foreign 

 boulders may be seen in the neighbourhood of Longridge Pell, 

 although it is possible that they may have been brought thither out 

 of the line by icebergs or coast-ice during the period of the Upper 

 Boulder-clay. At any rate the boundary will go over the western 

 end of the Rossendale anticlinal, and thence to Manchester. 



I wish particularly to call attention to the fact that this direction 

 coincides with that of the scratches, and that it is across tlie mouths 

 of all the valleys. 



In the part of the district east of Pendle Mr. W. Gunn's observations 

 confirm my own, to the effect that there are no boulders of other than 

 local rocks. He gives only one apparent exception — namely, the 

 existence on the north side of BouLworth Hill of blocks of quartz^ 

 rock, a rock which he had seen nowhere in place in the neighbour- 

 hood. A rock of this kind, however, I happen to know does occur 

 in Stoekdale, near Settle, 18 miles to the N.N.W. of Boulsworth ; so 

 that this is no real exception to the rule. Moreover the line of 

 transport coincides with the direction of the ice-sheet's movement as 

 shown by the scratches at high elevation on Bowland Knotts and on 

 Twiston Moor. The scratches at Kingsear in the neighboui^hood of the 

 parent rock are not in this direction, but S. 15° W. ; as, however, they 

 run along the side of some high scars, this variation is probably local. 



It is a point insisted on by some geologists, that wherever either 

 rounded stones or marine shells are found in the Boulder-clay 

 it must be of marine origin. I do not think that either of these sup- 

 posed characters is infallible. Mr. ' Croll has shown that the 

 Caithness TiU, which contains shells, need not necessarily be marine, 

 but may have been formed by the ice-sheet working over a previous 

 sea-bed and pushing the shells on to the land. In this way shells 

 scratched and broken maybe found at very miich higher levels than 

 the sea in which they lived and died. They are there as much boulders 

 as the scratched stones along side of them, and are no more evidence of 

 the drift in which they lie having been formed under the sea than Spi- 

 rifers and Producti found in Limestone Eiver-gravel would be proof 

 of its being marine. In very many places the ice-sheet must have 

 passed over what had previously been the sea-bed ; and if its course 

 took it thence inland we should be surprised not to find sea-shells 

 mixed with the drift formed by it. Nor can rolled stones be considered 

 a better test ; for under similar circumstances they would be brought 

 up from the old sea-bed on to the land. But rolled stones may be 

 found in abundance in the terminal moraine of any glacier, so that we 

 have their presence most naturally explained. No one who has seen 

 one of those swallow-holes called "moulins" in the ice of a large gla- 



