E. P. Cnherxcell — Theory of the Ice Age. 57 



This gives for F, the heat transferred to us from the Gulf Stream, etc., 

 in January, the equation 



F= 1455-265 = 1190 



Now, in order to get the direct effect of diminished sun-heat, let us 

 suppose that the heat transferred in January in the epoch of great 

 eccentricity is the same as that now transferred, and we get for the 

 temperature at that period the equation 



where S' is the reduced sun-heat radiation in the epoch of great 

 eccentricity. But the sun-heat is reduced by about 16 per cent., so 

 that if we neglect the star-heat we get for 8' the value 223. Hence 

 the equation becomes 



1190 + 223=^^^(460+0* 



and this gives for t a value between 36° F. and 37° F. Hence, on 

 this supposition the fall in temperature would be between 3° F. and 

 4° F., about the same as that obtained by my much safer method of 

 comparison. 



If we suppose that in July we receive a balance of heat from the 

 Gulf Stream and air-currents, we should get a still smaller lowering 

 of temperature in the supposed Glacial epoch. For instance, if R, 

 the July heat transferred, was equal to the July sun-heat, then the 

 above fall in temperature would be reduced by almost exactly one- 

 half If, on the other hand, we lose half our summer sun-heat by 

 transference to more northern latitudes, then the fall in temperature 

 would _ be about 7°, nearly double that obtained on the first 

 supposition. 



I do not give this as a safe method, because in the first place it 

 neglects the effect of the blanketing by the atmosphere ; in the 

 second place there is the uncertainty about E, and in the third place 

 although it is quite certain that the radiation increases faster than 

 the temperature, yet Stefan's law may not be strictly applicable to 

 the radiation of the earth in space. I insert it merely to show that 

 a little physical knowledge of the problem they were dealing with 

 would have saved the astronomical geologists from making such 

 extraordinarily mistaken estimates of the effect of the decreased 

 sun-heat on terrestrial temperature. 



The fact that the lowest temperatures occur almost simultaneously 

 about January all over the globe, as shown by the "Challenger" 

 isothermal maps (just as the highest temperatures occur about July), 

 indicates that the upper regions of the atmosphere have a very 

 great controlling influence on the loss of heat by radiation, and 

 that instead of considering that the earth radiates heat off to space 

 it would be more correct to regard it as radiating to those high 

 altitudes of atmosphere in which, for two reasons, the temperature 

 is probably much more uniform between the poles and the equator 



