SAODIES FOR STUDENES. 7 
for the drift. It has been seen that lakes could never have cov- 
ered the drift-mantled country, and therefore that the drift did 
not originate wholly or mainly through the agency of lakes. It 
has also been seen that the sea, acting alone, cannot account for 
the drift and its accompanying phenomena. It remains to inquire 
whether standing water, aided by the ice which forms, or which 
under favorable conditions may have formed on its surface, can 
account for the phenomena which water alone cannot explain. 
Since the distribution of the drift proves that it is not of lacus- 
trine origin, and since lake ice is restricted to lakes, it is clear 
that lake ice cannot afford the key to the explanation of the 
drift. Yet because the action of the lake ice is more familiar 
than that of sea ice, it will be profitable to study it briefly. 
The ice which forms upon the surface of lakes in winter may 
be subject to considerable movement both during the winter and 
in the spring. On fresh water, ice forms at a temperature of 
32° Fahr. The ice thickens as the isothermal plane of 32° sinks 
beneath the surface. Although water expands on becoming ice, 
the ice cannot suffer reduction of temperature without contrac- 
tion. If its temperature be lowered greatly, and that suddenly, 
as often happens in winter, the contraction of the ice is correspond- 
ingly great. The ice cover which fits a lake snugly at a temper- 
ature of 32° becomes too small, after contraction, to cover the 
whole surface of the water. It must either draw away from the 
shores, or crack. If the attachment of the ice to the shore is 
stronger than the cohesion of adjacent parts of the ice, the ice 
will crack. Otherwise it will draw away from the shore. Both 
withdrawal from the shore and cracking might take place in the 
same lake at the same time. The intense cold of the winter 
nights in temperate latitudes affords abundant illustration of both 
results. In whichever way they are formed, the breaches caused 
by the contraction of the ice will be healed by the freezing of 
the water which rises in them, and again the cover of ice fits 
snugly over the water beneath. When the temperature again rises, 
the ice expands. It was already large enough to cover the water, 
and expansion makes it too large. It therefore crowds the 
