SURFACES OF DISCONTINUITY 593 



known researches, found considerable support for the law pro- 

 vided M was taken to be the molecular weight of the liquid 

 and not of the vapor. Indeed this work was used to calculate 

 the degree of association in many liquids. Ramsay and 

 Shields actually made another slight modification of Eotvos' 

 law, writing it 



<i) 



(^ = ki^-^) (tc-t- d), 



where 5 is a small number, approximately 6. The "Eotvos 

 constant" k, they found to be 2.1. However, later research 

 has shown that the number k is not a constant for all liquids, 

 and that the use of this law as a method of measuring degree of 

 association is not reliable. Other suggestions have been made, 

 such as one by van der Waals based on thermodynamical 

 reasoning, viz., 



o-Q 



(■ ^ !)■ 



In this equation n is a constant for all liquids and 



o-o =A;(pc^O% 



where k is a constant for all liquids and pc and tc are critical 

 pressure and temperature. Experimental research shows that 

 this result also is not exact; although n for a number of 

 common organic liquids does not vary by more than a few per 

 cent from 1.21. Katayama (Set. Reports Tohoku Imp. Univ. 

 [1], 4, 373 (1916)) has suggested a modification of Eotvos' law 

 in the form 



•m 



^ = \~iir) ^^^ - ^^' 



where d is the density of the vapor; and actually an elimination 

 of tc — t from this and the equation suggested by van der 

 Waals, taking 7i to be 1.2, gives a relation 



a = C{D - dy 



