388 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 5 1 



Under this latter assumption all quantities relating to insolation 

 would be mathematically rigorous in contrast to those relating to 

 terrestrial radiation which can only be thought of as general average 

 values. 



However, the variations observed on the sun's surface make it 

 highly improbable that the intensity of solar radiation can be invari- 

 able, and therefore we must certainly also assume that the values 

 relative to insolation always have only the character of average 

 values. 



II. "The quantities of heat that are given or lost by a definite 

 portion of the earth's surface or the atmosphere in all the various 

 possible ways in the course of a year are on the average equal to 

 each other." 



This theorem is, like Theorem No. I, a direct consequence of the 

 assumption that the sun and the earth are in a certain sense in a 

 stationary condition, that is to say, that we are in general justified in 

 speaking of average values of the various quantities that come into 

 consideration. 



III. "The quantities of heat that individual portions of the 

 earth's surface or of the atmosphere gain by insolation or lose by 

 radiation, in the course of a year, are in general not equal to each 

 other, but there are portions of the earth where the insolation is in 

 excess and other portions where the radiation is in excess." 



The correctness of this theorem follows from the simple fact that 

 warm air and warm water flow continually from equatorial regions 

 poleward, while conversely cold air and cold water or ice flow from 

 polar regions toward the equatorial. 



Hence the equatorial belt continuously loses heat by convection 

 (and also certainly in the form of the kinetic energy of translatory 

 motion) which must be replaced by excess of insolation if the mean 

 temperature is to remain constant, whereas the reverse holds good 

 for the polar regions. 



We can therefore subdivide the whole earth into three zones, viz: 

 one equatorial in which the insolation exceeds the radiation, and 

 two polar, in which the radiation exceeds the insolation. 



These zones I will designate as "insolation zone" and "radiation 

 zone," respectively. 



The lines that separate these zones from each other at the bound- 

 ary surface of the atmosphere may be called "lines of equal insola- 



