Vel. 53.] GLACIER-LAKES IN THE CLEVELAND HILLS. 551 
to the impartial cutting of the river into a Drift-filled valley, 
although that leaves unexplained the coincidence of its invariable 
preference for the south bank. 
The three examples, Crunkley Gill, Sunny-Brake Slack, and 
Glaisdale, show, I believe, that an ice-obstruction extended across 
Eskdale at the same time that a free fall for water could be obtained 
at the levels which I have indicated, namely :—for Crunkley Gill, 
500 down to 450 feet; Sunny Brake, 475 down to below 400 feet ; 
and Glaisdale, about 350 down to 325 feet. 
Where, now, was the actual escape to the southward ? 
One of the peculiarities of the Fen-Bog overflow which, at the 
outset of this enquiry, convinced me of the significance of its features, 
is that it is the lowest outlet from Eskdale, saving only Whitby 
‘Harbour. The 500-foot contour is intact right along the southern 
watershed until Bay Ness is reached, half a mile from the coast at 
North Cheek, Robin Hood’s Bay. The 400-foot contour extends to 
about the same distance from the cliff-edge at Hawsker: and the 300- 
foot contour is unbroken up to within 100 yards of the cliff at 
Whitby High Light, 14 miles from the town. There is, moreover, 
no sign of an overflow-channel at any level from Bay Ness 
to Whitby Harbour ; at the latter place, however, there is a great 
rock-gorge 100 feet deep running along a line of fault. 
It is quite possible that the form of the ice-front was such that a 
broad way existed, a narrow marginal lake, across the space between 
Whitby Harbour and the High Light, but I have seen nothing to 
suggest it; and I think it more probable, taking into consideration 
facts presently to be adduced regarding the next stage of retreat, 
that the gorge at Whitby Harbour was the actual outlet. This, of 
course, need not imply that the coast at Whitby was free from the 
edge of the ice, for the cutting of the gorge may have been merely | 
commenced at this stage, and marine erosion on this coast has no 
doubt removed a broad strip, perhaps a mile or two in breadth, 
since the Glacial Period. But there must have been a free fall to the 
southward, and this can have been obtained only by low-level 
channels in a coast-tract since carried away, or by actual open 
water. 
The stages of retreat in Eskdale here described are indicated 
imperfectly by aseries of moraines. J have not succeeded in tracing 
any between the Lealholm moraine at Crunkley Gill and the series 
of successive ridges that descend from the northern moors across 
Eskdale at Glaisdale. These last differ in composition considerably 
from that at Lealholm, consisting mainly of sand and gravel, while 
the Lealholm moraine is almost entirely composed of Boulder-Clay. 
The difference is probably due to the latter having been accumu- 
lated wholly under water. 
I have already indicated that there are some signs that, at the 
period represented by the Wild-Slack and Crunkley-Gill overflows, 
a large volume of water was entering Lake Eskdale by way of Kildale 
and West Bank. For, just as a higher level of the lake is attested by 
the deltaic spread of gravel at Commondale, so a fragment of a lower 
