12S M. Bousslngavilt on Air found In the Pores of Snow. 



thesis of Dalton, who admits that the proportion of oxygen in the atmo- 

 sphere diminishes with its height. If, indeed, we consider snow as an 

 aggregation of small crystals of ice formed in the higher regions, avc 

 must necessarily conclude, on witnessing the large quantity of air which 

 it encloses, that, when the water dissolved in the atmosphere condenses 

 into snow, it does not expel this large portion of air which it always al- 

 lows to disengage when congealing on the surface of the earth ; if wo 

 may not suppose, say MM. de Humboldt and Gay-Lussac, that snow 

 retains a certain quantity of air enclosed in its minute crystals. 



Air adheres to snow in a very remarkable manner, and this shews that 

 it penetrates even the smallest crystals of ice. Very little gas is obtained 

 by passing snow under a bell-glass full of water at the temperature of 

 1° or 2°. The air is not disengaged in any abundance, except in the very 

 act of melting. This intimate penetration of the minute crystals which 

 constitute snow, cannot permit us to entertain much doubt that the air 

 derived from them comes for the most part from the regions of the at- 

 mosphere where the meteor is formed. According to the analysis which 

 I have given, we arc not entitled to believe that the composition of this 

 air is distinct from that of the lower regions, at least the difference, if 

 any such exist, is certainly of the nature of those which arise from errors 

 of observation. But, considered in relation to its origin, the air enclosed 

 in the interstices of snow presents sufficient interest to render its analysis 

 again desirable, when the processes of meteorological chemistry shall 

 have been suitably perfected. Uj) to the present day, however, it must 

 be understood that the results of experiment do not tend to confirm 

 Dalton's conjectures. Accordingly, in his memorable ascent, M. Gay- 

 Lussac having obtained air at the enormous elevation of 6680 metres, 

 did not find in it a different proportion of oxygen from that in the air of 

 Paris with which it was analyzed comparatively. In the work which 

 this celebrated natural philosopher prepared in conjunction with M. dc 

 Humboldt, he gives the oxygen of the air of Paris at 0°.21, and this 

 number scarcely differs from that deduced from the analysis made by M. 

 Brunner on the Faulhorn, at the height of 2600 metres, by a process 

 which certainly possesses advantages over the old method. M. Brunner 

 in fact found 20.915 for the oxygen of the air at this station. 



To complete, as far as it is in my power, the results obtained re- 

 specting the composition of the atmosphere at different heights, I shall 

 state the results of the analysis which I made during my stay among the 

 Andes. 



At Santa F^ de Bogota, at the height of 2643 metres, during the 

 month of April 1825, Volta's eudiometer gave me 20.65 for the oxygen 

 of the air. 



At Ibagut^ at the foot of the Quindiu Chain, height 1323 metres, I 

 obtained 20.7 for the oxygen of the atmosphere, on December 1826. 



At Mariquita, situate in the valley of the Rio-Grande de la Magdalena, 

 at an elevation of 548 metres, a series of anal^^ses by the spongy plati- 

 num, made in November 1826, indicated 20.77 as the oxygen of the air* 



