564 RICE ART. L 



all events, to the approximate character of the expression 

 ( — c/Rt) (da/dc) which was used for T calculated ; but in a later 

 paper with Wynne-Jones and Pollard (Coll. Symp. Monograph, 6, 

 57, (1928)) he abandons this explanation, as it was found for p- 

 toluidine that its partial vapor pressure over an aqueous solution 

 was directly proportional to the concentration of the solute. 

 This partial pressure gives a direct measure of the activity of 

 the dissolved p-toluidine and so there is no difference in value 

 between c(da/dc) and a{da/da). That being so, McBain 

 repeated still more decidedly a suggestion which he had already 

 made tentatively in the first paper, viz., that the situation is 

 complicated by the existence of surface electrification effects, 

 and that the omission of any consideration of these vitiates the 

 theoretical basis of the adsorption equation, as it stands, without 

 an additional differential term on the right-hand side represent- 

 ing increase in the energy of this surface electrification when 

 concentration increases by a differential amount. We cannot 

 deal with this point now, but will return to it at a later stage 

 of this commentary. A further point raised by McBain and 

 Davies (Jioc. cit.) is that in these and similar experiments "seldom 

 or never have true, two-component systems been actually under 

 observation, although this is fundamental. Solutions of 

 electrolytes or substances capable of hydrolysis, such as soap, 

 cannot be treated as two component systems except in the rare 

 event that the composition of the adsorbed material is identical 

 with that of the solute remaining in the solution." The point 

 of this remark is that we are implicitly using the equation 



da = —Tid/jLi — T2dn2 



and making Ti zero by adjusting the surface so that we have 



da = — r2(i) dn2. 



But this is invalid if there are still other components present. 

 As McBain and Davies say "The component (or components) 

 actually present, but hitherto ignored, is the gas (or air) in 

 presence of which the surface tension is measured when bubbles 

 are produced." If we set Ti for the solvent equal to zero there 



