14 Rev. EB. Hili—Eccentricity and Glacial Epochs. 
into space. Now, suppose that by some change in the distribution of 
the heat-supply, less heat is received in the cold season. Then more 
of the heat lost is radiated off in the formation of more snow. As 
much more heat is received during the warm season, just sufficient 
to melt the additional snow if wholly effective for that purpose. 
But since there is now more snow than before, more heat is reflected 
away, and therefore less than the whole of the additional heat of 
the summer is available for melting. Thus less than the whole of 
the snow formed in the winter will be melted in the summer, or, at 
all events, less heat will remain available to warm the earth in the 
Warm season, and accumulation of snow or of cold will begin. 
But while this cause alleged by Dr. Croll is theoretically correct, 
what practical amount of effect it may have is quite unknown. I 
am not acquainted with any researches into the proportions of heat 
reflected and absorbed by ice and snow. Hxperiments are much to 
be desired. The rapidity with which dark objects sink into snow, 
and small stones into glacier-ice, shows that ice and snow must reflect 
much heat. On the other hand, snow and ice, like other good 
reflectors, are bad radiators. They will not cool so fast and far— 
they will not radiate away so much heat-—as the ground would if 
left bare. Snow again is known to be a good non-conductor of heat: 
the young corn has been saved by it from many a frost, and the 
Esquimaux in their snow-hut may perspire with warmth. The 
coating of snow thus tends to diminish the winter’s loss of heat in a 
manner precisely answering to the way in which it diminishes the 
summer’s effect in melting. The actions of non-conduction and bad 
radiation are precisely contrary to that of reflexion, and for anything 
we know to the contrary may precisely balance it. 
Great stress is laid by Dr. Croll on the Third cause, the effect pro- 
duced by clouds and fogs in lowering temperature. I cannot think 
they have any effect of the kind. Mr. 8. V. Wood’s objection, that 
vapours give out in condensation as much heat as they absorbed in 
formation, seems to me to be fatal. Dr. Croll brings forward a mass 
of testimony which proves that many cold and icy regions are 
subject to thick fogs. But it proves nomore. None of the evidence 
shows whether the fogs produce the cold, or the cold gives rise to 
the fogs. The latter is as possible. We do not attribute the cold of 
mountain summits to the clouds which so often settle on the peaks. 
Let us consider the effect on temperature of the formation and dissi- 
pation of a cloud or a fog. We know that not only is heat being 
constantly poured on to the earth; it is also constantly streaming off 
from it into space. We suppose that at present the mean annual 
temperature of places is not on the whole changing. ‘Therefore, on 
the average of years the total annual receipt of heat must balance 
the total annual loss. Expenditure must equal income if there be 
neither debt nor saving. Suppose, then, winter ended, spring begin- 
ning, and the sun pouring heat upon the earth. The heat converts 
snow and ice into water, water into vapour, but this is chilled by 
the neighbouring snow into mist and fog, which intercept the sun’s 
genial beams, retarding their effect on the frozen mass. This is Dr. 
Ss 
