J 292 



>.() and \ice versa. Let iis now tlnallv form (/ r — ^ </to,, 



and let us first integrate along a circle r,, i= constant ; then the 

 |K)siiive values of / will lie multiplied by greater values of (5, — 5,)' 

 than the negati\e values, M) that the positive sign results. If we 

 had calcnlated />:: — />, we had multiplied bv (C, — ?,)*, so that then 

 negative values of / had been multiplied by a greater factor, and 

 I he negative sign would have resulted. 



In order to arrive at Uist at an estimation of the oi-der of mag- 

 nitude of (p;i — /)), we observe that: 



'do; a 



~ )• dx ihi dz — ~ (22) 



Or A ' 



J 



in which a represents the known quantity a of the equation of state 

 and .\' the number of molecules per molecular quantity. 



1 

 We further assume that in the factor the radius of the sphere 



'\ 

 of attraction of the molecules o mav be wi-itten foi- /•,, and that 



.... (5-i,)*-(C-CJ^ (1,-5,)' 



the influence ot the factors and -- wdl con- 



'1 '11 



sist in this that the values which would be obtained by an omission 

 of these factors, are multi|)lied by a moderate vabie /(. smaller 

 than 1 . 



Thus we liud, when we also take into account thai .V^-7'= RT: 



«I/3.T (a)' 1 



ƒ(-.- — 1) =: — ril L. 



'^^ ' id HT ' .V' o 



(21./) 



If we had calcnlated j), we shoidd also have found a term with 

 L' in the virial of the atti'active forces. If we call it //'. then: 



a 



/' -= i ^' Y 



so that: 



n>e — V avSTT a 1 , ,. 



p' 3«//?7' ' .V o 



As our purpose was oidy a rough estimation, we have taken in 



this: 



a = I and rii = 1 



a = 3.10^ 

 /r=10-" 

 = 5.10-« 



.V=r6.10" 



