23 
Now of these causes the first and third exclude each other. 
If fogs hinder the sun's rays from reaching the earth, they 
must also prevent the ice and snow from radiating heat away 
into empty space. The dull, cloudy surface above must 
receive and absorb all the heat of the summer sun, and can 
allow little heat to radiate into space, except at night ; even 
then much less than under a clear sky. Of course, till the ice 
and snow are all nearly melted, they effectually hinder a sensible 
rise of heat above 32° ; but this is only the converse of their 
previous effect, in their formation, to hinder a lowering of the 
temperature till the whole has been frozen. All the heat of 
the sun which falls on the earth must produce its full effect, 
either in raising the ice, snow, and the ground itself, up to the 
freezing-point, or in melting them, and turning them into 
water or aqueous vapour. The same amount of cold which 
would depress a stratum of chalk ten feet deep to the zero of 
Fahrenheit would spend itself in turning 7^ inches of rainfall 
into ice and snow. Thus the presence of moisture, whether 
in the air or the soil, or lakes and rivers, is the most effectual 
hinderance to excessive lowering of the winter temperatures, 
so long as the total annual heat received from the sun is not 
diminished. But in the imagined glacial epoch, this total 
amount is increased per cent, for the whole globe, and 
3 per cent, for the northern hemisphere. 
40. Even with the corrections before named, the calculation 
cannot lead to a precise result, but shows at the most a limit 
towards which the temperature would tend, if the solar heat 
and radiation into space maintained the given proportions for an 
indefinite period of time. If the rule were sound, some very 
unnatural conclusions would follow. Each pole, during its 
winter of half a year, when it receives no heat at all from the sun, 
would sink to the temperature of space, or — 239° F. Again 
the heat which the pole receives from the sun at midsummer, 
exceeds that received by an equal surface at the equator in the 
ratio of tv. sin. t to cos. «, or F3G38 to 1. But since the 
summer heat of the equator is 79°, or 318° above that of space, 
the midsummer heat of the pole, by Mr. Croll's mode of 
reckoning, should be 115° higher, or 194°, little short of the 
heat of boiling water. Each conclusion is plainly very wide 
of the truth. 
41. Again, Mr. Croll insists forcibly on the vast amount 
of heat transferred northward by the Gulf Stream. He reckons 
it equal to one-fourth part of the whole amount received from 
the sun by the Atlantic area or basin, from 25° N. up to the 
Arctic Circle. The consequent increase of the mean tempera- 
ture of Great Britain is not less, he thinks, than 30 c ; but in 
