418 Prof. G. N. Antonoff on Surface 



support the above view entirely. The works of Harkins, 

 Langmuir, and Svedberg indicate that the electromagnetic 

 forces play a considerable role in capillarity, and, according 

 to Chatley *, these forces may be both electromagnetic and 

 electrostatic, which would explain the complexity of vec- 

 tor ization. 



It is especially difficult to settle the question as to the law 

 of molecular action. In the previous paper it was selected 

 as the inverse fourth power law, which follows from the 

 assumption that the distance between the doublets exceeds 

 considerably the length of the doublets. If on the contrary 

 the distance between the doublets is small, the forces would 

 vary as the inverse fifth power of the distance (see Chatley, 

 loc. cit.). 



We may take as an example of this, a case described in 

 ' Molecular Physio,' by J. A. Crowther (1919). See figure 1. 



Fig. 1. 



• o 



o • 



The outer rings of electrons in two adjacent atoms are repre- 

 sented by the black dots, and the effect of positive electricity 

 is regarded as equivalent to that of the charges concentrated 

 in the white circles between them. If the atoms are brought 

 together ^ ery close, they will turn so that the electron of one 

 will face the positive charge of the other. In this case 

 apparently the attraction will be inversely proportional to 

 the fifth power of the distance. 



But whatever the actual law of molecular action, it is a 

 question of secondary importance so far as our chief results 

 are concerned. The scope of the paper was chiefly restricted 

 to the problem of interfacial tension All the calculations 

 were done assuming the inverse fourth power law. It can 

 easily be shown that the same would hold true whatever the 

 law of molecular action. 



In the previous paper it was shown that the following 

 equation must be true when two liquids are in equilibrium 

 one above another (case of partial solubility) : — 



P.-P^-Pi^ijV' 3 -^ 1 '' 3 , • • • (1) 



* Proc. Roy. Soc. Loud. xxxi. pt. 3, April ]5,p. 92 (1919). 



