SCIENTIFIC BALLOON ASCENSIONS VON BEZOLD 299 



approximately asymptotic to the adiabats of the dry stage, since, 

 for the remarkably slight moisture contents that correspond to the 

 highest parts of the curves, the "latent heat of condensation no 

 longer suffices to perform any appreciable fraction of the work of 

 expansion. 



Moreover, at the greatest altitudes the emission and absorption 

 can play only a very subordinate rdle on account of the extraordi- 

 nary rarity of the air, so that the changes in these strata take place 

 nearly adiabatically. 



We thus come to the very important result that at great altitudes 

 the temperature curves more and more nearly approximate to the 

 adiabats of the dry stage and therefore the vertical temperature 

 gradient must tend toward the value of i°C. fall per ioo meters 8 . 



The course of the temperature curve constructed as formerly 

 according to the numbers deduced from Glaisher's observations 

 must therefore from purely theoretical grounds seem very doubtful, 

 at least in the highest portions. The same is of course true of the 

 formulae of Hann (for moisture) and Mendelieff (for temperature) 

 which rest on these same observations. 



It may be said certainly that one of the most important attain- 

 ments of the undertaking described in this present publication 

 (Ergebnisse Wiss. Luftfahrten, etc.) is the fact that, as regards the 

 diminution of temperature in the highest strata of the atmosphere, 

 there has been established a complete accord between theory and 

 experience. 



3 Here it may be especially stated that all these remarks as to rising and 

 falling currents are only first theoretical approximations. In fact it will 

 frequently happen that masses of air that have ascended over very warm 

 places and with a relatively large humidity arrive overhead with such high 

 temperatures that it is impossible for them to descend at neighboriug local- 

 ities. Under such conditions abnormally warm air overhead must spread 

 horizontally and flow away to great distances above the lower strata of air. 

 Such phenomena as give rise to very slight vertical temperature gradients 

 and may in fact lead to temperature inversions have been frequently observed 

 in recent years even at moderate altitudes, as was, for instance, the case in 

 September, 1900, when such a warm layer extended from the Alps to the 

 North Sea (see W. Briickmann "Die, etc." The Temperature inversions 

 in summer anticyclones. Inaugural Dissertation. Berlin, 1904), whereas 

 it occurs as a regular phenomenon in the much higher strata and such cases 

 are certainly referable back to the air flowing out from the tropics (see Assmann 

 "Ueber, etc." "On the existence of a warm current of air at the altitude 

 of 10 to 15 kilometers." Sitz. Ber. d. Berlin Akad. 1902, pp. 495-504). 

 It cannot be too often stated that the considerations here set forth are only 

 crude approximations. (Note added in 1905. W. v. B.) 



