Carbonic Acid on Smooth Glass Surfaces. 171 



on smooth glass surfaces gives rise to, go far to support the 

 supposition that it is not due to simple adhesion, but that 

 we have to deal with a chemical decomposition of the glass, 

 and consequently that glass is a substance which might be 

 quite unfit for such investigations. Such a supposition, 

 however, loses its force when we take into consideration the 

 following facts: — That at the temperature we are considering 

 anhydrous carbon dioxide gas is an acid so weak that it 

 is not even capable of expelling the water contained in per- 

 fectly anhydrous potassium hydroxide; whilst, on the other 

 hand, silica possesses such a powerful affinity for the bases 

 present in glass, that it not only expels completely at white 

 heat the carbonic acid from potassium and sodium carbonates, 

 but also sets free the acid of these salts at lower temperatures 

 on boiling their aqueous solutions. 



That the chemical affinity of anhydrous carbonic acid 

 towards the anhydrous bases of glass should, in the absence of 

 water, be able to exceed that of silica, is the less plausible since 

 such a decomposition would be resisted by the mass-action of 

 the silica, which in the glass used amounted to almost five times 

 that of the acid contained in a neutral silicate. 



The only view that remains is, therefore, that the carbonic 

 acid itself as such, occurs on the surface of the glass without 

 having been transformed into a chemical compound. As a 

 solid or fluid mass solely under the influence of gravity be- 

 comes surrounded by an atmosphere of gas of diminishing 

 density, so must the accumulations of gas of diminishing den- 

 sity take place on the surface of solid or fluid bodies in con- 

 sequence of capillary attraction. According to Quincke's 

 researches this attraction can still be perceived up to a 

 distance of 0*000005 centim. from the surface; so that 5*135 

 cubic centim. of carbonic-acid gas at 0° C. and 0*76 metre pres- 

 sure, condensed in three years on a surface of 10,000 square 

 centim. of glass, must have occupied a space of 0*05 cub. 

 centim., and have undergone a condensation of at least from 

 5*135 to 0*05 cub. centim. Such a condensation corresponds, 

 however, to a pressure of 102*7 atmospheres. Now carbonic- 

 acid gas at the temperature of the experiments, 19° C, even 

 under a pressure of 57*5 atmospheres, is fluid, so there can be 

 no doubt that the surface of the glass was covered with a 

 layer of liquid carbonic acid. 



If the whole of the 5*135 cub. centim. of gas be present 

 in the fluid state, then, since the specific gravity of liquid 

 carbonic acid at 19° C. is 1*137, the thickness of this fluid 

 layer under a pressure of 1 atmosphere would only amount 

 to 0*00000888 millim. It is easy to understand, however, 



