ART. L 



552 RICE 



Thus to obtain the same surface concentration we require for 

 each successively higher member of the series a bulk concentra- 

 tion about one third of that of the previous member, and so the 

 higher members are more and more "capillary active," to use a 

 common term which designates the property of causing a lower- 

 ing of surface tension and being in consequence adsorbed in 

 excess quantity at the surface. It will be observed that, if c is 

 large compared to a, Szyszkowski's formula approximates to that 

 of Milner. A relation has just been found from the former, viz., 



where g is a constant at given temperature and would be in fact 

 the upper limiting value of r if the law held for extremely high 

 concentrations. Now this relation is virtually equivalent to an 

 equation deduced by Langmuir {J. Am. Chem. Soc, 38, 2221, 

 (1916)) for the adsorption of gases on a solid surface 

 (plane crystalline). Although not of special interest now, it 

 may not be amiss to mdicate Langmuir's argument in broad 

 fashion, inasmuch as Gibbs at a later point in his treatment 

 deals with the conditions at a surface separating a solid from a 

 fluid. 



Langmuir's special hypothesis is that the molecules of the 

 gas are "condensed" on the crystalline surface when they strike 

 it, and do not in fact rebound in an elastic fashion as sometimes 

 postulated in kinetic theory of gases, except in a minority of 

 impacts. There is a good deal of evidence that this is actually 

 the case, and that in general the molecules remain on the 

 surface for a longer or shorter time depending on the attractive 

 forces between the solid and the adsorbed layer, and on the tem- 

 perature. There is therefore a concentration of molecules on 

 the surface whose amount depends on the average length of time 

 during which the molecules remain upon it. This state of 

 affairs obviously resembles what happens when molecules of a 

 solute pass from the solution into the surface layer and so it 

 is not surprising that there should be a formal resemblance in 

 the laws deduced in the two cases. Indeed Langmuir's analysis 

 could be easily adapted to give a theoretical foundation for 



