E. Andrews on Human Antiquities at Abbeville, de. 183 
the latter to leave a vacancy when it melted. An examination 
of the gray gravel adjacent confirms the idea of powerful ice 
action. It consists of chalk flints, mixed with broken chalk 
of every size, froma fine powder up to fragments as large 
as a man’s head. Many of the fragments, though soft 
enough to write with upon a blackboard, have preserved with 
absolute perfection the sharp angles and edges which they had 
at the time they were broken from the cretaceous strata. It 
does not seem possible that they could have been rolled a hun- 
dred feet in the bed of a stream without losing this sharpness. 
It follows that much of the material of those beds was either 
dro from floating ice, or else deposited by the mechanical 
action of ice fields floating down the ancient river, which crushed 
the edges of chalk strata abutting on the valley, and pushed 
the debris along to: be left wherever the irregularities of the 
channel permitted. The agency of ice is further evinced by 
the occasional presence of large boulders of sandstone in the 
gravel, some of which weigh a ton. These must have been 
transported from far up the stream, as the rocks in the vicinity 
are exclusively chalk. The crushing action of ice against the 
edges of the valley is rendered more probable by the fact that 
the angular mixed gravel is found only in the border beds, 
while the strata nearer the center of the valley are washed com- 
paratively clean of chalk. : 
The envelopment of ice and frozen gravel in sedimentary 
beds is by no means a phenomenon peculiar to the valley of the 
Somme. In a former article, I described the angular masses 
bsg clean gravel which occur in the drift clay beneath the 
melt. 
The evidence of powerful force exhibited in the gravel beds 
of Abbeville and Amiens renders untenable Sir John Lubbock’s 
