07i the Diffusion of Gases, 8fc. 77 



defiant gas did, although the diffusion of the former gases was 

 more opposed by mechanical causes. 



It is evident that the diffusiveness of the gases is inversely as 

 some function of their density — apparently the square root of 

 their density. 



f The results, however, are much influenced by the mechanical 

 resistance arising from gravity, which is not constant in gases 

 of different densities, the position of the receiver remaining the 

 same. The effect of the position of the receiver may be con- 

 ceived from an experiment on hydrogen gas. The re- ^-x 

 ceiver, filled with hydrogen gas, was placed in an upright 

 instead of a horizontal position (see figure). Other 

 circumstances being the same as in the experiment of 

 table (1), of 150 parts hydrogen 22.1 were found re- 

 maining in the receiver after diffusion for ten hours, 

 instead of 8.3 parts, as in that experiment. 



Although the stoppers fitted precisely, the additional 

 precaution of luting the joinings was attended to. The 

 properties of the receiver, too, were found not to be 

 peculiar to it. 





II. — On the Diffusion of mixed Gases into atmospheric Air. 



In the case of an intimate mixture of two gases, I was 

 anxious to learn if each gas left the receiver, independently of 

 the other, in the proportion of its individual diffusiveness — 

 which would be a step gained in the solution of the important 

 problem of the analysis of mixed gases by mechanical means. 



For this purpose, the receiver was filled with 75 vols, hy- 

 drogen + 75 vols, defiant gas, agitated and allowed to stand 

 over water for 24 hours, that the mixture might be as perfect 

 as possible. The receiver being then placed in the usual 

 position, the mixed gases were allowed to diffuse into the air 

 for ten hours. The receiver thereafter was found to contain 



Hydrogen gas . . . 3.5 

 Olefiant gas . . .56.6 

 Air 89.9 



150.0 



