802 REPORTS ON THE STATE OF SCIENCE.—1912. 
III. and [V.—INFLUENCE oF NatuRE oF SOLVENT AND SOLUTE. 
For many years the attention of investigators of solubility 
phenomena was devoted almost exclusively to the study of cases in 
which water was the solvent used. A surprising amount of work was 
done which dealt with aqueous solutions, and the opinion grew up 
that water as a solvent possessed quite unique properties and was 
quite distinct from other solvents. It was, however, becoming 
generally recognised that solubility in some way involved the mutual 
interaction of solvent and solute. 
With the advent of the ‘ionic dissociation hypothesis’ and its 
application to solution, this belief in a special property of water became 
much strengthened, mainly from the fact that by making certain 
assumptions it was found possible to treat the problems presented by 
solutions mathematically, a possibility which was freely exercised by 
physical chemists. Notwithstanding much evidence to the contrary, 
one of the assumptions made was that the solvent played no part 
whatever in conditioning solubility, and that therefore the problems 
could be treated precisely as if the solute diffused into a volume of an 
inert gas. 
The need of modifying this hypothesis was soon manifest; the 
ionised portion of the solute was then regarded as being in combination 
with the solvent. This modification having proved to be insufficient 
as an explanation of the facts elicited by a study of the problems on 
broader lines, it was agreed by the promoters of the hypothesis that 
not only the ionised portion but also the non-ionised portion of the 
solute must be in combination with the solvent. 
Now that reciprocal interaction between solvent and solute is a 
generally accepted proposition as explaining the facts, it becomes a 
matter of first importance to study solubility phenomena in relation 
to the properties, both physical and chemical, of the substances 
concerned. 
A considerable amount of work having direct bearing on this aspect 
of the subject has been published; and, following the plan adopted in 
Part I. of this Report, the subject matter has been classified in this 
Section III. according as it relates to the influence of: A, the physical 
nature of the solvent; B, the chemical nature of the solvent. 
In Section IV. A and B a similar classification is adopted with 
respect to the solute, and an additional subsection C has been included 
to bring together work relating to solubility as a function of atomic 
and ip Ba weights. 
III. A.—Physical. 
1896 Arctowski1* measured the solubility of sulphur, bromine, 
* jodine, mercuric iodide, and tin tetraiodide in carbon bisulphide, 
and of iodine in benzene, chloroform, and ether at temperatures close 
to the solidifying point of the solvents. The results led to the con- 
clusion that the solubility graphs may be prolonged beyond the point 
of solidification of the solvent, and that this point has no peculiar 
significance. 
Arguing from previous work on the velocity with which solid 
