Earth's Superficial Temperature. 23 



glacier receives no permanent superficial increase. Below this 

 limit the destructive begin to prevail over the productive agencies. 

 The distance to which the glacier descends below it depends on 

 local circumstances ; but we find that nearly ail glaciers, large 

 enough to be considered of the first order, descend to levels lower 

 than the snow-line by an amount varying from about 4000 to 

 5000 feet. In smaller glaciers the descent is proportionally less. 

 Again, the snow- line bears certain relations to the line which I 

 have defined as the line of 32° F., or that along which the mean 

 annual temperature is equal to 32° F. At certain places in suffi- 

 ciently high latitudes this line will lie at the level of the sea. In 

 lower latitudes it will lie at higher levels, and in still higher 

 latitudes the mean annual temperature will be less than 32°. It 

 is found that at the equator the snow-line lies about 1000 feet 

 below the line of 32°, while in the higher north latitudes it lies 

 above' the latter line, the vertical distance between them being 

 very variable. A continental climate, in which the atmosphere 

 contains comparatively little moisture, and the variation from 

 summer to winter temperature is very great, is favourable to a 

 relatively high position of the snow-line ; while an insular climate, 

 in which the quantity of moisture is comparatively large, and the 

 annual variation of temperature comparatively small, superinduces 

 a relatively low position of this line. Thus in the north-eastern 

 part of Asia the snow-line is probably from 4000 to 6000 feet 

 above the line of 32°, while in Iceland its height above the latter 

 line does not exceed a few hundred feet. A knowledge of these 

 facts enables us to estimate approximately the vertical distance 

 between these lines in any proposed hypothetical case. To esti- 

 mate the absolute height of the snow-line above the sea-level, we 

 have only then to calculate the height of the line of 32° at the 

 place proposed. For this purpose we must estimate the mean 

 annual temperature there by means of the isothermal line passing 

 through the place, and then calculate the vertical height to which 

 we must ascend to reach the point at which the mean annual 

 temperature is equal to 32° ; and to effect this we must know the 

 height which corresponds to a decrease of temperature of 1°. 

 Humboldt and others have shewn that this height may be taken 

 as varying from about 320 to 350 feet in ascending steep moun- 

 tains, or vertically in a balloon ; but Humboldt has also shewn, 

 from his own observations, that, in an ascent presenting a suc- 

 cession of high and extensive table-lands, the increase of height 

 for each degree* may amount to 450 or 500 feet. This is an im- 

 portant distinction. 



In this manner, then, the height of the snow-line above the 

 sea level can be estimated at any proposed place, with any hypo- 

 thetical distribution of land and sea. If a mountain rise a few 

 hundred feet at least above the snow-line, and the configuration 



