( 429 ) 



This xaltK', loo, is lower lluiii tluit coitipiiled befoi'c on other 



— vb" 

 supjjositions ; now 0,30, before 0,40. But lor the relation — y- 



0,396 0,295 



about Ihe same value is touiul ; before — = 5,2 ; now = 5,5, 



0,076 0,0534 



so only siig-hlly highei". 



Finally wo tind for hi- \ hi, according 1o (13): 



' = 1 — 0,0453 X 0,700 = 1 — 0,0317 = 1 . 



hr, 31,5 



A 6 Ab: bk 

 For as we found the value 0,628 for , we i^et = 0,628, 



vk — b]. ' vk 



h~ 



Lh lib 

 and hence — = -- = 0,628 X 1,114 =: 0,700. 



So reniarkablv enoneli we tind for the same value as van der 



30 1 

 Waals o-ave tor it, viz. — = 1 . 



31 31 



The above may already suffice to establish the conviction that the 



ordinary tlieory of association in itself gives a function of v for b 



(through the quantity /^), which already on the supposition ,i' = l, 



i.e. association to iknihL' molecules on an average, represents the 



ci-ilical data pi'i-fecth/ conwct/i/. ( )f the said double molecules only 



0,045 = ' .^3 are still |)resent in this form : "V-j, has already broken 



up into simple molecules. And yet in consequence of this Vk : b/- 



falls from 3 to 2,1, (i from 0,375 to 0,265, / rises from 4 to 7, 



hi, 'SO — vb" ' S a 



while z= , — ^^ — ^5, and the factors t\ and f\ of— i-- and 



bg ol b ' ' 27 6 



la. 

 -—^^ in the expressions for R7\ and p/^. are both very near unity. 



5. When ,i' i= 2 instead of ./'=:1, i.e. r = 3 (association to triple 

 molecules), we find from : 



m 8 l+.c 



- = „ and fi = 0,947 — ^— , 



n 7 l+.^•i3 



calculated from ƒ = 7 and f< =: 0,265 (see I, p. 295), in connection 

 with the expressions (5) (loc. oil.) for m and 7i : 



/■?— 0.9.^84 ; r/o-r 0.9158. 



29- 



