THE REVOLUTIONS OF THE CRUST OF THE EARTH. 339 



amount. If, on the contrary, under the same circumstances^ the winter 

 coincided with the perihelion, the earth would then be 4,529,784 leagues 

 (over 15,000,000 miles) nearer the sun during the winter than in sum 

 mer, and consequently the difference between the two seasons would be 

 almost annihilated in our latitudes. But as the winter of one hemis- 

 phere corresponds with the summer of another, it results that while one 

 hemisphere is subjected to the alternate influence of extremes of heat 

 and cold, the other, on the contrary, enjoys a perpetual spring-time, or, 

 we may better say, a uniform climate. 



It is true that, according to J. Herschel,* the quantity of heat the earth 

 receives from the sun is proportional to the angle the earth describes 

 in passing over any part whatever of its orbit, or, which amounts to the 

 same thing, the sum of heat received between the vernal and autumnal 

 equinoxes is constant for two parts of the year, whatever may be the 

 eccentricity of the orbit. It results, then, that the greater heat of the 

 summers of one hemisphere are exactly compensated for by their short 

 duration, while the summers of the opposite hemisphere, notwithstand- 

 ing the great distance from the sun, receive, on account of their length, 

 the same amount of heat as the warm summers of short duration. The 

 same comparison may be applied to the winters of the two hemispheres. 

 These considerations induced Herschel to deny all influence of the ec- 

 centricity and of the precession of the equinoxes upon the distribution 

 of solar heat upon the surface of the globe. And this is the cause of 

 the refusal of many geologists to recognize the connection which exists 

 between these cosmical phenomena and the variations of temperature 

 which have taken place in geological periods. 



But Humboldt observed that the temperature of the globe depended 

 as much upon the quantity of heat radiated into celestial space as upon 

 the heat proceeding directly from the sun, and the principle of compen- 

 sation might be applied to the latter but not to the former. It is in fact 

 well known that the temperature of a place rises whenever the days are 

 longer than the nights, and falls, on the contrary, when the nights are 

 the longer.t Now the austral hemisphere has not only a more rigorous 

 climate, on account of the greater distance of the earth from the sun, 

 but the winter is also longer, and the number of its nights greater. 

 The length of the winter, the small quantity of heat received from the 

 sun, and the heat lost by radiation and not compensated for, contribute 

 to cool the hemisphere to such a point that the humidity produced by 

 the evaporation of the waters in the tropical regions is principally pre- 

 cipitated upon this hemisphere, which, by its coldness, acts as a con- 

 denser. The result is an accumulation of snow and ice in the temperate 

 and cold regions, which prevents the action of the solar heat by absorb- 

 ing a large portion for their fusion. 



The direct action of the sun is also prevented by the precipitation of 



* Transactions Geolog. Soc. Lond., 2d ser., vol. iii, p. 298. 

 +J. Adh6mar, Revolution de la Mer, 2d edit., 1860, p. 16. 



