46 Correspondence—Bernard Smith. 
relating to the composition, character, and location of the glacial 
drifts, as well as that furnished by erratics, ice-moulded surfaces, 
and strie. 
Having thus divorced these anomalous channels from their 
surroundings, he has constructed a theory to explain their origin 
(as it seems to me) based chiefly upon their shape and rate of fall. 
If land-ice ‘‘ occupied all this district during some part of the Ice 
Age’’, and there was no Irish Sea Ice, I venture to ask why (as its 
markings show) the ice from Eskdale first turned south, along the 
seaward slope of Black Combe, and then swung to the east and north- 
east into the mouth of the Duddon estuary; and why the Whicham 
Valley Ice, moving first in a south-westerly direction, was also 
directed towards the south-east and east near the mouth of that 
valley? Why did it not ride out to sea? What was the impelling 
force that turned it aside ? 
Professor Bonney doubts whether marginal streams could cut 
channels in granite in a short time. He may read of recent examples 
described by Von Engeln,' to whose article I have already made 
reference. 
My critic admits that the systems of parallel trenches on the west 
coast, as contrasted with the preglacial drainage, are abnormal in 
direction, and finds some difficulty in explaining this anomaly, but 
confesses himself happier when dealing with the channels east of 
Black Combe, where ‘‘the trenches take a more normal course’’. 
The explanation issimple: in the first case the trenches were marginal 
to the Irish Sea Ice, and are therefore transverse to the normal 
drainage, whereas in the second case they were marginal to the local 
| Lake District glaciers which occupied the present drainage lines at 
a late stage of the glaciation. If, as Professor Bonney maintains, the 
dry channels on the west coast are preglacial, how does he explain 
the presence of thick glacial drift upon the ground between them, and 
upon the higher inland slopes, but not within them ? 
Perhaps the weakest part of this new hypothesis is an attempt to 
explain the ‘in-and-out’ channels by marine erosion of the seaward 
wall. A preglacial submergence cannot be invoked, because one of 
the long ‘in-and-out’ channels (Monk Foss) is cut entirely in glacial 
drift; nor can we admit a postglacial submergence, for that would 
have entirely destroyed the typically hummocky character of the 
drift on the plain between Millom and the mouth of the Esk. 
Moreover, even were ‘in-and-out’ channels at this spot due to marine 
erosion, we cannot explain in this manner the ‘ in-and-out’ channels 
of other districts far from the sea. 
Finally, one might pertinently ask how pre-Triassic valleys could 
have maintained such sharp well-defined contours to this day. As 
a field-geologist I have frequently noticed the “‘ half effaced features 
of an earlier topography ’’, but in few instances have I seen anything 
more blatantly modern in appearance than these marginal or overflow- 
channels in the Black Combe district, or, indeed, in North Wales, 
where they are cut in fairly soft shales. 
1 “Phenomena associated with Glacial Drainage and Wastage’’: Zeit. 
Gletscherkunde, vol. vi, pp. 126-31, 1911. 
