the Constitution of mixed Gases. . 193 



experiment, with all the attention I could bestow, this did 

 not prove to be the fact : for similar measures of impreg- 

 nated water gave up equal bulks of carbonic acid to like 

 quantities of all the different gases. 



The converse of this fact also occurred to me in the 

 course of a series of experiments, to which I have already 

 referred; viz. that the admixture of common air with car- 

 bonic acid gas diminishes considerably the proportion of 

 the latter gas taken up by water. Thus, when 20 measures 

 of pure carbonic acid gas are agitated with 10 of water, at 

 least 10 measures of gas are absorbed. But from a mix-' 

 ture of 20 measures of carbonic acid with 10 of common 

 air, 10 parts of water take only six of carbonic acid. That 

 chemical affinity between the mixed gases is not the cause 

 of the diminished amount of absorption, is perfectly clear ; 

 since it is indifferent, as to the effect, what gas is added, 

 and the proportion alone influences the result. The effect is 

 therefore to be ascribed to the diminished density of the 

 superincumbent carbonic acid by mixture with another gas ; 

 and the pressure of gases being directly as their density, and 

 the quantity absorbed by water being as the pressure, the 

 absorbed carbonic acid must necessarily quit the water. 

 This escape continues till the carbonic acid above the water 

 has a density equal to that in the water, and no longer. 



Previously to my acquiescence in your theory of mixed 

 gases, I undertook an extensive series of experiments, with 

 a view to ascertain the order of affinities of gases for water. 

 But, after a great variety of trials, made with all the accu- 

 racy in my power, I could discover nothing like a series of 

 elective attractions. Each gas, it was found, displaced every 

 other, and, reciprocally, was dislodged by them. 



Tt may be urged against the doctrine of the non-gravita- 

 tion of gases on each other, that from water, impregnated 

 with carbonic acid gas, and exposed to the atmosphere, the 

 gas ought, on this principle, to escape as rapidly as under 

 an exhausted receiver. It must be remembered, however, 

 that the escaping gas constitutes, by admixture with the air 

 of the atmosphere, a gas of diminished density, but still of 

 such density as to retard the escape of further portions. 

 All that the air-pump effects is to remove these as fast as 

 they are liberated. 



There are various facts satisfactorily explained on this 

 doctrine which are irreconcileable to any former hypothesis. 

 Of these I shall mention only a few; since the theory will 

 receive from yourself all the elucidation that its establishment 

 can require. 



1st, If each gas be a vacuum to every other, a heavier 

 4 gas 



