SURFACES OF DISCONTINUITY 509 



The analogy between the quantity a- for an interface between 

 two Kquids or between a Hquid and a gas, and the quantity T 

 tor a membrane in tension between two gases, is thus drawn 

 once more from another standpoint. It is therefore quite 

 natural for Gibbs at this point to say, as he does, that equation 

 1 499] or [500] "has evidently the same form as if a membrane 

 without rigidity and having a tension a, uniform in all directions, 

 existed at the dividing surface," and thereupon to suggest the 

 name "surface of tension" for a specially selected position of 

 the dividing surface and the name "superficial tension" for cr. 

 The cautious nature of Gibbs' statement might easily be over- 

 looked by the reader. It clearly does not commit him to the 

 view that the interface between two fluid masses must be 

 regarded actually as a membrane in a state of tension. This 

 idea is certainly a prevalent one, and the treatment of "surface 

 tension" in many of the elementary texts of physics fosters it. 

 So it may be of some service to the reader if a short discussion 

 of this much debated point is inserted here. This will require 

 us to enter into a more detailed consideration of the molecular 

 structure of the fluid phases than actually occurs in the original, 

 but that is hardly avoidable in any case in view of the develop- 

 ments of Gibbs' work by subsequent writers. In addition, later 

 workers have availed themselves of the statistical calculations 

 and results which are nowadays associated with molecular 

 pictures of matter in order to give a deeper interpretation to 

 some of Gibbs' results and to help to elucidate certain difficulties 

 of the purely thermodynamical treatment. So it may prove 

 serviceable to seize the opportunity at this point to give also a 

 brief discussion of the fundamental statistical idea involved in 

 such calculations. 



II. Surface Tension 



3. Intrinsic Pressure and Cohesion in a Liquid 



The behavior of soap films, in which there may well be a 

 strong lateral attraction between long-chain molecules such as 

 those of the fatty acids, "anchored," as it were, side by side in 

 the surfaces of the film (an attraction which may with some 



