Dr. Croll's Hypotheses on Geological Climates. 225 



equator? The reason why such convection-currents do not 

 arise is, that the normal state of the air is a stable equilibrium, 

 even a very stable one, so that if, by any cause, the higher 

 strata be forced down, they would acquire, by compression, a 

 much higher temperature than the lower have. 



An unstable equilibrium, a few cases excepted*, exists only 

 in the lower strata up to a few thousand feet, and this only in 

 the day-time, when the surface of the ground is much heated 

 by the sun. It disappears about sunset, or even somewhat 

 earlier. It does not occur in winter in high or even in 

 higher middle latitudes (say, from 50° onward). All this has 

 been established by so many observations in mountain-coun- 

 tries and in balloons that it can scarcely be doubted. 



In any case, if Dr. Croll doubts such well-known facts, i. e. 

 if he doubts that between the lower strata of air and those 

 some tens of thousands of feet high there exists a stable equi- 

 librium, the onus probandi rests with him. In fact, a tem- 

 perature of— 50° F. must occur at such a considerable height 

 above sea-level, that the air, if forced down, would arrive 

 warmer, and not colder, than the air existing at sea-level. 



Besides, I maintain that the radiation of particles of air is 

 but a very trifling cause of loss of heat by our globe, and that 

 by far the principal causes of it are the radiation of the sur- 

 face of the ground (or snow) and that of the surface of the 

 water. If the lowering of temperature of the surface of land 

 and of the air be considered, the loss of heat by radiation of 

 the surface of the ground (and snow) is by far the most im- 

 portant. 



I can much better agree with Dr. Croll in what he remarks 

 about the conservative character of snow once formed, though, 

 as will be seen further on, I do not agree as to the import- 

 ance of winter in aphelion during high excentricity. Besides, 

 I think Dr. Croll does not well understand the cause of the 

 fogs, which have been so often noticed in high latitudes in 

 summer. He thinks the fogs are caused directly by the 

 melting of the snow, and, interposing a screen between the 

 sun and snow, are effectual in lessening the amount melted. 



The melting of the snow by the sun has not the power to 

 cause fogs by itself. On extensive continental regions of 

 Europe, Asia, and America snow lies in winter, and is melted 

 from March to June, yet fogs are of exceedingly rare occur- 

 rence during that time, and they are more frequent at night, 

 i. e. are rather instrumental in preventing the loss of heat by 

 radiation than the heating influence of the sun's rays. In 



* For example, in thunder-storms, hail-storms, &c. 



