568 Dr. R. D. Kleeman on Molecular Attraction 



Similarly we have for the surface tension \ the equation 



X = 



Xah 



where 



{.U \.Wfii-T)..5 S 0HJLV. . (2) 



\jl,f,U,Vj ^ \X C l c / Z? 2 V ' 



*i 3 =^aW|tw + iu) 2 + w 2 +V}, 

 and x ab =(-^-—) > and -i ■*' f> as before, 



denotes an operator between given limits. This equation 

 follows from an investigation given in the same paper as 

 that on the heat of evaporation. 



The various forms of the arbitrary function that satisfy 

 the above equations must be found by trial : we have already 

 referred to some examples at the beginning of the paper. 

 But it will be easily seen that we can give the function 

 almost any form we please provided its real values are not 

 restricted to lie between certain limits, and it contains a 

 sufficiently large number of terms or independent constants. 

 Further, the equations enable us to determine the nature of 

 the constants in any empirical relation of the heat of 

 evaporation or surface tension with other quantities, by giving 

 the arbitrary function a form that gives the relation required 

 and comparing the constants obtained with those contained 

 in the empirical relation. Examples of this will be given in 

 this paper. 



Surface Tension. 

 If we put 



in the formula for the surface tension, we obtain 



where a l5 « 1? /?i . . . , are constants which are the same for 

 all liquids. An equation of the form \ = A(1 + TB) has 

 already been used by physicists to represent the surface 

 tension, and found to agree approximately with the facts. 

 The former of the above equations indicates the nature of 

 the constants involved. 



Before proceeding it will be of importance to consider a 

 point in connexion with testing the truth of relations deduced 

 theoretically. It may happen — and in most cases does 



