CONTORTED STRATA IN DRIFT. 
179 
Fig. 115. 
Gravel and a 
sand. 
right of/*, g. The overlying coarse gravel and sand, is in 
some places horizontal, in others it exhibits cross bedding, 
and does not partake of the disturbances which the strata 
c, have undergone. The underlying till is exposed for a 
depth of about 20 feet; and we may infer from sections in 
the neighborhood that it is considerably thicker. 
In some cases I have seen fragments of stratified clays and 
sands, bent in like manner, in the middle of a great mass of 
till. Mr. Trimmer has suggested, in explanation of such phe¬ 
nomena, the intercalation in the glacial period of large irreg¬ 
ular masses of snow or ice between layers of sand and grav¬ 
el. Some of the cliffs near Behring’s Straits, in which the 
remains of elephants occur, consist of ice mixed with mud 
and stones; and Middendorf describes the occurrence in Si¬ 
beria of masses of ice, found at various depths from the sur¬ 
face after digging through drift. Whenever the intercala¬ 
tion of snow and ice with drift, whether stratified or unstrat¬ 
ified, has taken place, the melting of the ice will cause such 
a failure of support as may give rise to flexures, and some¬ 
times to the most complicated foldings. But in many cases 
the strata may have been bent and deranged by the mechan¬ 
ical pressure of an advancing glacier^ or by the sideway 
thrust of huge islands of ice running aground against sand¬ 
banks ; in which case, the position of the beds forming the 
foundation of the banks may not be at all disturbed by the 
shock. 
There are indeed many signs in Scotland of the action of 
floating ice, as might have been expected where proofs of 
submergence in the Glacial Period are not wanting. Among 
these are the occurrence of large erratic blocks, frequent¬ 
ly in clusters at or near the tops of hills or ridges, places 
which may have formed islets or shallows in the sea where 
floating ice would mostly ground and discharge its cargo on 
