ACTION OF TERNARY ARRANGEMENTS. 27 



ume of oxygen, equilibrium would not be obtained until a certain amount of carbonic 

 acid was found in the gas, and a certain amount of oxygen in the water. And the 

 same would hold in the case of any other gases or any other liquid. In the course of 

 experiment, examples of this case are often met with. The water commonly used in 

 pneumatic troughs contains both oxygen and nitrogen. If into a jar containing such 

 water we pass pure nitrogen, in the course of a few minutes oxygen will leave the wa- 

 ter to diffuse itself into the nitrogen. Had we thrown in pure oxygen, nitrogen, on the 

 contrary, would have deserted the water and mingled with the oxygen gas. In gaseous 

 analysis, this action, which obtains to a greater or less extent with every gas, often gives 

 rise to much perplexity. 



> 78. TERNARY ARRANGEMENTS. It is plain that the conditions of the action consid- 

 ered in the last paragraph may be obtained at once by suitable arrangements ; and, as 

 it is important that these should be well understood, I shall dwell upon them minutely. 



79. In paragraph (77), we considered the reaction ensuing, first, of a single couple 

 or binary arrangement, and then the disturbance effected by the introduction of another 

 element. Could we, then, at once have exposed the volume of water by one surface to 

 oxygen gas, and by another to carbonic acid, the changes that were consecutive would 

 have been simultaneous. Let a (fig. 13, pi. 1) be a sheet of water, on which, at its 

 upper surface, a volume, b, of carbonic acid reposes, and beneath its under surface, c, a 

 volume of oxygen ; both gases pass at once through the water, in opposite directions, into 

 each other. It is evident that the thinner we make the barrier of water, the more rap- 

 idly will equilibrium be obtained. This I have accomplished in the following manner, 

 by using mere liquid films, and for that purpose have taken advantage of soap and other 

 bubbles. A glass tube \ inch in the bore, and seven or eight inches long, is to be drawn 

 out at one extremity to a capillary termination, and when the bubble is to be blown, the 

 other end is dipped into a solution of soap. The tube having been previously passed 

 through a cork, as in Jig. 14, pi. 1, is now to be introduced into a clear vial or bell 

 glass, the neck of which the cork fits loosely; on blowing at the capillary termination, 

 the bubble slowly expands in the vial, where it is protected from access of air. To 

 measure its diameter, I take a strip of white pasteboard, and divide it into inches and 

 decimals, placing it in such a position before the vial that it crosses the bubble diamet- 

 rically; then with a small telescope that magnifies twelve or twenty times, and at the 

 distance of about eight feet, I observe the bubble much magnified, the micrometrical 

 pasteboard apparently passing through its very substance, as is shown in the figure. 



80. Through a soap bubble T53 inch in diameter, the substance of which, previous 

 to expansion, was contained in a cylinder 1 inch in diameter and \ in height, ammonia, 

 either pure or diluted with atmospheric air, passes instantaneously when air from the 

 lungs is on the other side. Into the bottle in which the bubble is to be blown a little 

 strong solution of ammonia is to be poured ; the bubble is then expanded ; at a particular 

 point it becomes dyed with the richest hues, and that moment the phenomenon of 

 endosmosis is complete : care must be had to suffer no moisture from the mouth to 

 close the capillary termination of the glass tube ; and now a rod, a, dipped in muriatic 

 acid, is to be brought over the opening ; as the bubble is collapsing by the attraction of 



