536 MR. P. F. KENDALL ON A SYSTEM OF [Aug. 1902, 
From Sneaton Low Moor the watershed runs in a general easterly 
direction, and remains intact and unbroken quite to the coast : there- 
fore the retreat of the ice must have been in such wise that no lakes 
were formed, for neither direct overflows, nor any sign of marginal 
overflows can be found. 
There are many ways by which this fact may be explained. The 
retreat may have been so rapid that, though evanescent lakes were 
produced, the overflows were not cut deep enough to be recognizable : 
several shallow grooves can, it is true, be seen to cross the watershed, 
as on the Whitby road. Or, again, the shrinkage of the ice may have 
been progressive from east to west, so that no lakes could form. Or, 
as a third explanation, it is possible that marine denudation has 
removed a coastal region which was channelled by overflows. 
A combination of the first and second of these suggestions seems 
to me to explain the phenomena most fully. The third hypothesis 
appears to be excluded by the form of the ground at Bay Ness, where 
the watershed reaches the coast. There are, however, as will beshown 
later, clear cases in which seaward tracts carrying overflows have 
been wholly or partly swept away. 
(7) The Eastern Coastal Tract. 
The line of watershed extending from the head of Iburndale round 
Sneaton Low Moor marks the beginning of a drainage-system dis- 
playing many remarkable anomalies, some of which have been ably 
described and elucidated by Mr. Fox-Strangways. A narrow coast- 
strip of country, extending from Robin Hood’s Bay on the north to 
Hunmanby on the south, and varying in breadth from 100 yards 
up to a maximum of about 3 miles, drains in a general way down 
the normal slope of the land into the sea. But, behind this, at a 
short distance, seldom more than 3, and never more than 6 miles, 
there runs a great gorge or connected series of gorges (like the 
intercepting drains of the engineer) which receives all the drainage 
of the hinterland and carries it away to the southward, and finally 
westward, through the Vale of Pickering, into the Ouse and Humber 
drainage. ‘Thus it is that streams, rising within 2 miles of the sea 
at Robin Hood’s Bay or Peak, pass into a system which enters the 
sea at Spurn. . 
In the initiation of this drainage, the effects of an ice-sheet 
which shut the seaward ends of the valleys is clearly traceable, 
just as was the case in Northern Cleveland, but with this difference, 
that in the northern area the lake-overflows were rarely cut to so 
great a depth as permanently to deflect the larger drainage-channels, 
On the eastern coast, the ice-invasion was not so extensive, 
and the overflows were all of the marginal type. Some of the 
cutting was over high and prominent watersheds, but more often 
existing drainage-lines were followed and deepened. Moreover, the 
effects were cumulative, as one aligned sequence remained in 
occupation for a long period, and an increasing volume of water was 
brought to bear upon just those regions where the greatest barriers 
