CONDITIONS OF PRESENT GLACIATION DISCUSSED. 323 



receive a large enough ainount of precipitation, as we have seen to be the 

 case in the Scandinavian Range. There the high and comparatively level 

 summit has in places large areas of permanent snow or neve ; and if the 

 precipitation were considerably increased, no doubt these areas would be 

 correspondingly enlarged, for the mean temperature is sufficientl}' low to 

 cause all the precipitation to take the form of snow, over a much larger part 

 of the range than is now permanently covered ; on\y, as we have seen,* the 

 heights of land nearer the ocean from which the moisture-bearing wind 

 comes cut oflf the precipitation, which is not large enough to hold out so as 

 to reach more than a very short distance inland.f 



Reasoning from analogy, we may therefore suppose that if a low level 

 area was cold enough to receive its precipitation in the form of snow, and at 

 the same time was so situated that the winter snow-foil should be larger than 

 could be melted away by the summer's sun, there would lie an accumulation 

 of snow or ice. Which of the two would finally cover the country the 

 prevailing climatic conditions must determine. Here a few words may prop- 

 erly be introduced in regaixl to the conversion of snow into ice, since that is 

 a process which directly connects itself with the question before us. 



Ice may be and is formed in nature in two ways: either by direct freezing 

 of water, as in the case of the ordinary- congelation of the supei'ficial water 

 of rivers, lakes, or ponds, or even of the ocean surface; or by the transforma- 

 tion which snow undergoes when exposed to suitable conditions after it has 

 fallen upon the ground. The writer's long and careful examination of snow 

 and ice masses in various climates and positions leads him to the conclusion 

 that in nature ice is hardly formed at all from snow except through the inter- 

 vention of water. This water may be derived directly from rain-fall, or it 

 may be furnished by the rapid melting of the superficial layers of snow\ In 

 some regions, as for instance in the Alps, it is the rain-fell which is chiefly 

 concerned in bringing about the transformation in question ; in other coun- 

 tries it is — at least in large part — effected by the saturation of the mass by 

 the water derived from its own meltine. That this must be the case seems 

 to be clearly- proved by the manner in which the dry snow of the upper 

 regions becomes changed first into a slightly granular mass, then into decided 

 nece, and finally into well-characterized glacier ice. The position of the neie 



* See ante, pji. 288, 289. 



t The difference in the ainount of jireoifiitation on the two slopes of tlie Scandinaviiin Range is very great. 

 See Kniuimers Raiii-ehart of Europ-, in Zeitsehrift der Gesellschaft fur Erdkuu(]e zu Bi-rlin, Band XIII. 1878. 



