394 
flE VIEW. 
whether by the action of currents and winds, or of pro- 
truding headlands, must present throughout its entire 
area a varying momentum and resistance. This, in 
connection with the fact of the hummock ridges or 
lines of junction being the soonest to give way, will 
explain the facility with which this great pack yields 
to assailing forces from without. 
I believe I have adverted already to another most 
interesting and beautiful provision of nature to prevent 
the reconsolidation of the ice after it has been once 
broken up during the seasons of thaw. Fragmentary 
masses, which were fast cemented during the winter 
to the under surface of the floe, now rise through the 
water, interposing themselves between the opening 
tables, and acting as checks or wedges to prevent their 
reapposition and cementation. 
By such impressive compensations does nature ef- 
fect the equilibrium of the year. In a short and irreg- 
ularly-graduated season, this great ice-raft, the growth 
of nine months of congelation, is returned to water by 
means almost independent of thaw, and resumes its 
office of tempering the climates of the distant south. 
As the views I have detailed in this chapter of the 
causes which effect the final disintegration of the pack 
may perhaps be novel, I venture to recite them in the 
form of a summary. 
First. Changes in the molecular condition of the 
ice at temperatures below the freezing point, giving 
rise to infiltration of salt water and rapid decomposi- 
tion of the ice in consequence. 
Second. A greater intensity of this action, owing 
to the infraposition and superposition of two fluids of 
differing densities, inducing a rapid circulation allied 
to endosmosis. 
