( 437 ) 



Remembering, that n^^n^, a^^=: a^^, the last term maybe 

 shnplilied to 



2 



For f*'y^-,— f*V/, ^^^ *^"^^ ^^ similar e\|)ressioii. In this, however, 

 the quaiitilies l\, /•., (Ihe lieal-capacities of the same Ions, at inliiiite 

 volume) and {c.^^, (/%)„, (ï^,,),,, (ï/,)„ (llie energy and Q^\\vi)\)y -constant 

 of those Ions) will be exactly the same, a^^, a^^, a^^imd <i^^ will ulso 

 remain unaltered, so that for the diiference (u',. — m' ) — in ~n' ) 



vr^^ ^C7i ^^^ /v.j *■ CI./ 



we may write : 



RT 



^.J 





"l(«-,l-«-.l) ^''l(«'8l-«'4l) 



F 



^^' 



') 



+ 



The qnantities, relating to the second sohent, ai'o indicated by 

 accents. 



We may now go a step furlliei- and accept as a first ai)i)roximation: 



^'3 = ^'4 ' ^'\ = ^'\ 1 ('?.Z = f'4 4- 



If we then also w^rite 



F F' , n 



— =: V , ~j- ■= V , //.^ rrr: // (1 — a) ^ — ^:rz c , etc., 

 1 ^' 1 ^'1 



we iinall}' obtain for A 

 1 



L = 



31 -41 "31 -41 A I . ./'(I—"')''' (1— «K' 



V V J \ V V 



(G/>) 



As, in conscqneiice of the e(|uilil»riuni of partition, is 



(1 — a)c 



conslani, A will ha\c the form 



A = A„ + ;i(i-.0'^-, 



or since, on acconnt of the equilibi'iiini ol dissocialioii, = 



(1— «)c' 



constant, also the form 



A = A., + A'(«.r. 



Whether A \vill he posilixc or i)egati\e, depciids rliiclly on A„. If 



< 



A will l)e j)()siti\'e. We also see, that A- A^, \\ ill iuci'case or doci'case 

 with {\\G scco/k/ powci' of t< c, that is to say in the case of sironglv 

 dissociated electrolytes, where « is nearly 1, almost with (•"■'. 

 Dec. 1902, 



