VOLUME IN SOLUTION 43 



The change due to taking the volume in solution into 

 account is thus in all cases a small one ; still, it gives 

 a somewhat better result 1 from the fact that similar 

 changes in composition are accompanied by somewhat 

 more corresponding changes in volume. Anyhow, the gain 

 is not very great, since for oxygen, e.g. it is necessary to 

 assume four volumes, according as it appears in the form 

 of carbonyl, or ether, or isolated hydroxyl-oxygen, or 

 hydroxyl in the neighbourhood of another oxygen atom, in 

 the carboxyl group. Taking these constitutive influences 

 into account, an empirical formula like that on p. 40 may 

 be developed for the volume in solution, and, like the 

 other, may be made use of for molecular weight determina- 

 tions. But there is hardly any need of this, seeing that 

 reliable osmotic methods exist for the case in question. 



Let us next discuss what has been arrived at with regard 

 to electrolytes. Here additive relations were first dis- 

 covered by Bender and Valson, but were expressed in 

 terms of the density. Later calculations refer to the 

 volume in solution. 



The most essential fact is that whilst the volume of 

 non-electrolytes, as shown above, varies little with the 

 concentration, that of electrolytes varies largely 2 : 



Normality NaCl (18) \ H 2 S0 4 (18) 



00025 6-93 



0-005 16.39 7.94 



0-05 16-37 I2 '77 



o-i 1657 14-05 



i 17.97 16.96 



5 20-9 18.52 



only becoming constant when the maximum molecular 

 conductivity is reached, as is the case for sodium chloride 



in normal solution, but for sulphuric acid only on much 



greater dilution. Strictly additive relations are only to be 

 expected for these limiting values, and do actually occur 



1 Traube, 1. c. 



2 Kohlrausch and Hallwachs, Wied. Ann. 53. 14 ; 56. 185. 



