396 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 51 



The dates of these extremes lead to a division of the year into 

 seasons of warming and cooling, which division opens up other 

 points of view than the division of the year on a purely astronomical 

 basis. 



The geographical distribution of the total energy over the surface 

 of the earth gives for the first time an idea of the actual distribution 

 of heat at the earth's surface, whereas hitherto we have given this 

 name to what is actually only the distribution of temperature in 

 the lowest stratum of air. 



But we must first come to an understanding as to a zero point 

 if we wish to compare together the energies of different points of 

 the earth's surface, although this is a matter that need not at all 

 be considered in the investigation of the annual or the diurnal 

 changes at any special points. But this is a point on which I will 

 dwell in a later communication. 



At present I need only remark that the approximate computa- 

 tion of the total energy in the sense just established ought not to 

 encounter insurmountable difficulties. 



In fact the portion belonging to the solid earth's crust is deter- 

 mined with relative ease, as will be shown in the second part of the 

 present communication. 



The extraordinary importance of the solution of this problem may 

 be seen from the remarks that will be made when we speak of the 

 remarkable variations that the so-called average temperature of 

 the whole earth (i. e., of the lowest stratum of air over the whole 

 earth) experiences in the course of the annual period, whereby it 

 will be found that it is not allowable to reason from this directly to 

 variations of the total energy of the whole earth. Similarly the 

 importance of this question will appear in considering the peculiar 

 behavior that the polar regions show at the time of the maximum 

 altitude of the sun. 



V. "The quantities of heat that enter and leave through the 

 outer limit of the whole atmosphere in the course of a definite 

 portion of the year are not necessarily equal to each other." 



If the earth's surface and the atmosphere were perfectly homo- 

 geneous, at least throughout each surrounding layer concentric 

 with the earth's center, and if moreover the earth's orbit about the 

 sun were circular, then the above-mentioned equality would neces- 

 sarily exist, but since these conditions are not fulfilled, and since 

 the regions of excess of insolation and excess of terrestrial radiation 

 change their locations [on the globe] in the course of the year and 



