288 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 51 



sidered, are questions that can only be determined by observations 

 in balloons. 



Moreover the other fundamental question, in what manner the 

 ascent and descent of air takes place in the areas of high pressure 

 and low pressure can only be explained in this way. For there can 

 be no doubt that these movements are by no means so simple as 

 are represented in the sketches in the text-books, but that hori- 

 zontal and vertical motions are combined together in the most varied 

 and intricate manner and that mixtures etc., take place. 



Those questions also that belong to pure dynamics, in distinction 

 from the just mentioned thermodynamic questions, can only be 

 specifically considered with the help of research by means of balloons. 



The great difficulty of this problem forbids its general treatment as 

 a whole; we must therefore consider the individual portions sepa- 

 rately and afterwards attempt to establish the connection of these. 



Among the results attained for which we have to thank the pres- 

 ent series of balloon voyages (by the Berlin branch of the German 

 association for promotion of aeronautics) the first place must be 

 given to the elucidation that they have given us as to the warm- 

 ing and cooling of the atmosphere and the general distribution of 

 temperature and moisture in the vertical direction. 



(2.) THE VERTICAL DISTRIBUTION OF TEMPERATURE FROM A 

 THEORETICAL POINT OF VIEW 



In a memoir recently published (No. IX, pp. 216-228, or No. 

 XIII of these translations) but whose principal contents I communi- 

 cated to the Berlin Academy of Sciences on May 5, 1898, I tried to 

 state in a purely theoretical manner the influence of adiabatic ascend- 

 ing and descending currents of air on the average distribution of 

 heat in the atmosphere. 



I started with the assumption that radiation inward or outward 

 can only be influential at the earth's surface and at the upper sur- 

 face of the clouds, and that a gain or loss of heat in the free cloud- 

 less atmosphere by absorption or emission can only play a subordi- 

 nate part and may be neglected in our final approximations. 



It seems that these assumptions do actually suffice to explain 

 at least the prominent features of the vertical gradient of tempera- 

 ture, although further elaborations are needed in respect to many 

 peculiarities. 



At the same time these considerations led to views relative to the 

 exchange of heat in the atmosphere that had indeed been indicated 

 by several investigators, especially by Lord Kelvin and by H. von 



