﻿2 Mr. William Sutherland on the 



and directly proportional to the product of the parameters 

 V3Am characteristic of each. Although the parameter is 

 written in the form \/3Am apparently involving the mass m 

 it will be shown that \/3Am is independent of the mass m 

 but is a function of the size of the molecule ; it would there- 

 fore be better denoted by a single symbol a, so that the 

 attraction between any two molecules M x and M 2 is a^/r 4 , or 

 between two M 1 is a^/V 4 , the parameter a being a function of 

 the size of the molecule but not directly of its mass. Thus, 

 with Gm 1 ?7? 2 /r 2 to denote the gravitational attraction of two 

 molecules M. 1 and M 2 , the general expression for the force 

 between them is 



Qm^n^/r 2 + a x a 2 h A . 



The dependence of the coefficient of diffusion of two gases 

 on the attraction between their molecules w T as indicated in 

 general terms in a recent paper on the Viscosity of Gases and 

 Molecular Force (Phil. Mag., Dec. 1893). In that paper it 

 was shown that in those parts of the kinetic theory of gases 

 which depend on the number of encounters of a molecule per 

 •second (or, in other words, on its mean free path), the effect of 

 molecular force cannot be neglected as of only secondary 

 importance ; it is fundamental. Thus the complete expres- 

 sion for the coefficient of diffusion of two gases will involve 

 the attractions between their molecules in a manner now to 

 be established ; but as the kinetic theory of the diffusion of 

 gases, even when simplified by treating the molecules as 

 forceless, is in a little confusion (there being at least three 

 forms of expression for the diffusion-coefficient in the field), 

 it may be desirable to recapitulate briefly the theories of the 

 diffusion of forceless molecules from the three points of view. 



The first in time is that of Stefan, accepted by Maxwell ; 

 the second is 0. E. Meyer's, given in his book on the ' Kinetic 

 Theory of Gases ;' and the third is that of Tait (Trans. Boy. 

 Soc. Edin. xxxiii.), who has treated the diffusion of gases 

 rather elaborately. 



Stefan's theory is this : — If two gases are diffusing into one 

 another, then at any point one has a general velocity a x in 

 one direction, and the other a velocity a 2 in the other, the 

 density of the first diminishes in the direction of a l3 of the 

 second in that of a 2 . Consider, then, an element of the first 

 of section unity and length 8x in the direction of a x : the 

 partial pressure due to its molecules at one end is p 1} and at 

 the other p x + $x dp^^/dx, so that there is a driving pressure 

 ScV dp 1 /dx which is resisted by a resistance like friction offered 

 by the other gas in the length &e, which may be denoted by 



