1874.] Climate of the Glacial Period. 429 
year were much more in one hemisphere than the other, it 
is quite certain that more heat would be radiated during the 
nights that were longest. But, and this is the fallacy on 
which it seems to me Mr. Croll’s argument rests, the earth 
radiates heat in the day time as well as at night, and this 
has not been taken into consideration. The warmth of 
the day depends on the excess of heat received over what 
is radiated, not that there is no radiation at that time; and 
if we take into account the heat radiated during the day we 
shall find that no more is lost in one hemisphere than the 
other from that cause. And if the absolute amount of heat 
received from the sun be equal whatever the amount of 
ellipticity, and the absolute amount of loss. by radiation 
also equal when we calculate that radiated during the day 
as well as that during the night, it is evident that the 
absolute difference between the heat received and the heat 
lost, or the mean temperature of the two hemispheres due 
to these causes alone, must be the same whatever the 
amount of ellipticity of the orbit may be. 
and. Would the lengthening of the ice-accumulating 
period and the shortening of the ice-melting period cause a 
greater accretion of ice? Here again Mr. Croll and his 
followers answer unhesitatingly in the affirmative, and they 
put it in this way :—‘“‘ At the time of greatest eccentricity 
during the long winter of aphelion, longer by thirty-six 
days than the summer of perihelion, such an accumulation 
of snow and ice would have taken place that even the 
diminished distance between the earth and the sun in 
summer time would be powerless to effect its removal.’’* 
Here, again, I think the argument is based on a miscon- 
ception. It is not a fact that our winter begins as soon as 
the sun has passed the autumnal equinox, though what 
is called the winter solstice does. The nights are longer 
than the days, but snow does not immediately begin to fall 
nor water to freeze, and our winter does not commence on 
the 22nd of September, but several weeks later. In the 
shorter but hotter summer of perihelion some excess of 
heat must be stored up in the earth, the sea, and the 
atmosphere, not to be entirely given up until long after the 
winter solstice has been entered on. ‘The advocates of this 
theory affirm that the mean temperature would be lowered 
because the heat of the short summer would be taken up 
in melting the ice that had accumulated in winter, but a 
pound of water in passing from a liquid to a solid state 
* The Great Ice Age. By JAMES GEIKIE. 1874. P. 139. 
VOL. IV. (N.S.) 31 
