ON THE CONTRACTION AND EXPANSION OF ICE. A419 
ef an inch to an inch, and in the distance of a mile the number 
counted exceeded one hundred. 
These fissures were, of course, the effect of contraction, and their 
aggregate widths fully compensated for the absence of the larger 
fissures, which I expected to have seen, and were quite equal to the 
' Maximum amount of expansion witnessed on any one occasion. 
Thus, then, was I convinced of the error of my previous statement 
regarding the contraction of ice, and actual measurements since made 
have fully proved that the expansion equals the contraction of ice for 
equal changes of temperature. 
The cause of ice contracting in the peculiar manner above men- 
tioned, is owing, of course, to its unequal thickness, glariness, and 
density ; the shrinkage bemg unequal throughout the mass, the lines 
of fracture are accordingly numerous and irregular. 
‘Were ice equally thick, dense, and glare, it would contract uniformly 
towards its centre, and we would not then witness those irregular 
cracks and fissures just described ; neither would be seen that en- 
largement, or piling up of fractured ice, which is the effect of expan- 
sion under ordinary circumstances. 
Ice at formation is at its greatest or maximum dimensions, and. 
although the temperature of water may be far below 329, the latent 
heat given out during crystallization (as is well known) will instantly 
raise the temperature of the ice to that figure. 
The formation of ice, like that of other substances, takes place at 
a certain fixed temperature, which is also that of its melting, and 
which remains constant during the process of its solidification. 
The first movement in ice, therefore, after its formation, must 
necessarily be shrinkage or contraction; the fissures which occur on 
a large field during this process immediately fill with water, which is 
soon frozen. The field ice, be it remembered, still extends to its 
original limits, and is in a state of shrinkage. 
Now, should the temperature rise to 32°, this ice will expand and 
overlap its original boundary by a distance just equivalent to the 
aggregate widths of the various cracks le previously represented 
the amount of contraction. 
This revivifymg or replenishing process accounts in the most satis- 
factory manner for the seemingly exhaustless expansive power of ice. 
If we take the maximum expansion of ice, at any one point on 
Rice Lake, and divide the amount by the radius or diameter of the 
