78 Professor Percy Kendall—Glacier Lake Channels, 
contours during the great lapse of time demanded for the excavation 
of the wide and very deep valley that is presumed to have usurped 
their functions. 
The preservation of the contours of these deserted trenches is, next 
to their anomalous relations to the relief, their most marked 
characteristic, and it has furnished the strongest argument yet 
advanced for the brevity of post-Glacial time. 
Whether excavated in granite, volcanic rock, slate, grit, sandstone, 
conglomerate, limestone, shale, or glacial materials, they show in a 
large proportion of cases scarcely any appreciable signs of weathering; 
the salient upper edges are only slightly guttered and the re-entrants 
at the foot of the ‘ batter’ only rarely are filled in by the running 
down of the banks. Exceptions to this rule are, however, not 
negligible, neither are they without significance. There are in 
Yorkshire, as is well known, several boulder-clays distinguishable by 
colour, superposition, and contents; they have, as I hope to show 
some day more fully, a geographical distribution of great interest: 
(1) the lowest, the Basement Clay, is not recognizable far from the 
coast, though representatives probably exist in many inland situations ; 
(2) the Purple Clay, represented in the country west of the Chalk 
Wolds of Lincolnshire by the Chalky Boulder-clay ; and (3) the Hessle 
Clay. The last appears to be limited to the seaboard of Yorkshire 
and the Vale of York, but an equivalent stage cf glaciation is 
recognizable in the valleys draining the eastern slopes of the Pennines 
from Swaledale to Airedale. 
A long interval marked by widespread and drastic denudation of 
the earlier deposits intervened between the Hessle stage and that 
which preceded it, and it seems probable that a great and general 
retreat and readvance of the ice took place in this interval. The 
maximum extension of the ice at the Hessle stage was many miles, 
and in some areas even scores of miles, short of that attained by the 
ice of the Chalky Boulder-clay, besides which, in the coast region the 
direction of its onset was different, the north to south component 
being preponderant, whereas at the earlier stage the impulse seems 
to have been directed more from the north-east. 
The limits of this readvance in Yorkshire can usually be traced 
with precision and clearness, in some places by well-defined moraines, 
in others by lake channels. The correlation of these two opposite 
classes of phenomena, the one an effect of deposition, the other of 
erosion, is comparatively easy—each forms the approximate boundary 
between a region in which the glacial deposits form an almost 
continuous mantle with the characteristic topographic features due to 
glaciation very fresh and complete, and areas in which the glacial 
deposits are reduced to a series of shreds and patches, often with long 
intervening spaces of driftless country. It is interesting to note that 
just as the Vale of York is spanned by the two great terminal 
moraines at York and Escrick respectively, so along the lower slopes 
of the Pennines two principal sets of channels trench the spurs 
between the river valleys. The upper set of channels is much 
shallower than the lower, whence it may be inferred that the extreme 
extension of the ice was rather ‘‘touch and go”’, and that a relatively 
