Intelligence and Miscellaneous Articles. 237 



carried far enough, not only all liquids, but solids also, such as salts, 

 would finally be reduced at ordinary temperature to the gaseous 

 state, as takes place under the action of heat. Finally, at a little 

 higher temperature (M. Kundt made his experiments at one tem- 

 perature only), cohesion being diminished by the increase of tem- 

 perature, the decrease of surface-tension under the influence of 

 pressure would take place still more quickly. 



Having recently determined the solubility of carbonic acid in 

 water under pressures of from 1 to 30 atmospheres, I proposed to 

 myself to make evident and establish the correlation of these two 

 classes of phenomena. Reserving the description of the method 

 employed and the numerical data for a special memoir, I will only 

 enunciate here the results of my experiments. 



Under pressures of 1-30 atmospheres, there exists a remarkable 

 relation between the laws of the solubility of carbonic acid in water 

 and the surface-tension of that liquid. That relation can be expressed 

 thus : — 



1. The product obtained by multiplying the surface-tension a by 

 tlie pressure P under which the carbonic acid is placed is proportional 

 to the saturation-coefficient S corresponding to that pressure — that is, 



aP=AS, 

 where A is a coefficient which depends on the temperature and in- 

 creases with it. 



According to the first law of the solubility, the temperature remain- 

 ing constant, _ decreases in proportion as the pressure increases*. 

 Experiment shows that the decrease of a, is proportioned to that of 

 p. With the aid of this relation of the phenomena of capillarity, 



those of the solubility of the gas can be calculated, and conversely. 



2. The pressure remaining constant and -equal ton atmospheres (n 

 being greater than 1), it follows from the laws of solubility that the 



(§L. 



quotient q. — decreases with the lowering of the temperature. 



(p)p=, 

 Experiment shows that in this case the ratio of the tensions cor- 



responding to these pressures decreases also. 



This result is in evident contradiction of M. Ivundt's theory, 

 since the lowering of the temperature, instead of retarding the 

 decrease of the surface-tension, accelerates it. The phenomena are 

 therefore completel3 r independent of the pressure, and depend only 

 on the state of saturation of the surface of the liquid — that is to 

 say, on the quantity of gas dissolved in the surface-layer. 



The above relation does not end at the pressure of 30 atmo- 

 spheres. The solubility increasing less quickly than the pressure, 

 tends towards a certain limit, which at 0° seems to be readied at 



* See my Note, Cotnptes Rendus, t. xciv. p. 1355. 



