EXPLORATIONS OF THE UPPER ATMOSPHERE. 313 



Still, perfect tightness lias only been obtained by a perfect adapta- 

 tion of the movable sleeve and a quite close fitting- of the cock which 

 at first stopped the effect of the train of clockwork. It was necessary 

 to add to the square piece terminating the sleeve of the cock a toothed 

 sector upon which engages the pinion of a barrel containing a spring 

 wound up until it effected the turning of the cock. The apparatus 

 being found to work well in the labaratory could then be used for 

 experimenting in inaccessible regions. 



M. Muntz, member of the institute, who undertook the analysis of 

 this air, obtained the following results : 



Volumes. 



;n 20.79 



Nitrogen 78.27 



Argon 0. 94 



The volume of the air obtained at 15,500 meters, at the temperature 

 of zero C, and at a pressure of 760 millimeters was 1.18581 liters. 



Composition of the air. — The amount of carbonic acid in 100 volumes 

 of air is 0.033 volume. 



In each 100 volumes of air deprived of carbonic acid the relation of 

 the argon to the sum of the nitrogen and argon is 0.01185. 



M. Th. Schloesing, jr., has determined tlie volume of the argon, by 

 the very exact method which he has invented, and has likewise super- 

 vised the analysis that determined the amount of oxygen. 



M. Cailletet, nevertheless, thinks that it wiH be necessary in the 

 future to gather anew quantities of air and submit them to compara- 

 tive analyses which will allow us to determine with certainty the com- 

 position of the air that fills these elevated regions, where scientific 

 instruments had never before penetrated. 



M. Muntz adds, in his communication to the Academy of Sciences, 

 that he has shown that the receiving apparatus worked in an irreproach- 

 able manner and that the cock would maintain the vacuum indefinitely. 

 There is no fear that other air will be introduced into the reservoir. 



The quantity of air contained in the reservoir may then be used to 

 calculate the barometric pressure of the altitude at which the air was 

 taken. The results of the analysis show already, what was indeed to 

 be expected, that at the altitude attained the composition of the air 

 does not differ much from that of the lower regions. 



But these figures must be accepted with some reserve. It will, in 

 fact, be necessary to still further improve the method of obtaining the 

 air so as to avoid all possible alteration in its comj^osition. For oiling 

 the cock it will be necessary to use a mineral oil incapable of absorbing 

 a trace of oxygen and consequently emitting a trace of carbonic acid, 

 at least under the conditions under which the experiment is made. It 

 will likewise be. necessary to use a reservoir whose walls can not absorb 

 any trace of oxygen. For this purpose a glass receiver would be an 

 ideal one, but a reservoir of gilded copper would seem to answer the 

 purpose equally well. In the present case we may suppose that the 



