VI: ON SOLUBILITY. 843 
carbon bisulphide with benzene and toluene the solvent powers of the 
mixtures were found to be greater than would be expected on the above 
assumption. 
The distribution coefficient of sulphur dioxide between water and 
chloroform was found by McCrae and Wilson }* to vary with the con- 
1903 centration, This fact they attributed to the varying degree of 
* electrolytic dissociation of the sulphurous acid and cited the 
experimental fact that hydrogen chloride diminishes the distribution co- 
efficient as being in support of their suggestion. Results not in har- 
mony with this view were published by Farmer **' who found the 
coefficient of distribution of benzoic acid between benzene and water 
to be but slightly changed by the addition of either potassium benzoate 
or sodium acetate to the aqueous layer. 
Vaubel '§’ found that in certain cases the distribution coefficient of a 
substance between two solvents is a function of the relative volumes of 
the latter. This was found to be true in the case of phenol dissolving in 
water and benzene or chloroform; m-cresol dissolving in water and 
ether; aniline dissolving in water and ether or chloroform, &c. With 
the system resorcinol dissolving in water and ether or benzene, however, 
the coefficient was observed to be independent of volume changes. 
1904 Vezes and Mouline *°* observed aqueous alcohol to be much 
* less soluble in turpentine oil than is that oil soluble in aqueous 
alcohol at the same temperature. 
Hudson '** made the important observation that the complete water- 
nicotine solubility curve is a closed one; above 210° and below 64° the 
two liquids are completely miscible: at about 90° the two layers have 
equal density: above 90° the nicotine forms the upper layer and below 
90° it forms the lower layer. 
The distribution of a number of acids and bases between water and 
amylic alcohol was studied by Herz and Fischer.1** These authors 
worked with acids such as phenol, acetic, oxalic and picrie acids, and 
bases such as ammonia, methylamine and triethylamine. Generally 
speaking the results obtained indicate that for weak electrolytes the dis- 
tribution ratio is independent of the concentration, but that with more 
highly dissociated compounds, for example, oxalic and picric acids, a 
constant is obtained only for the undissociated portion. 
1905 The liquid-pair diethylamine and water were found by 
’ Lattey 27 to be completely miscible only below 143.5°C. Above 
that temperature two layers are formed. 
The distribution of soluble substances between water and various 
aromatic hydrocarbons has been studied by Herz and Fischer *°® who 
found that a constant value for the distribution ratio is obtained when 
allowance is made for (i) electrolytic dissociation in water ; (ii) associa- 
tion in the hydrocarbon. 
906 Riedel 2*° observed that the partition coefficient of aniline be- 
are. tween water and toluene varies with the concentration, and as an 
explanation of this fact, suggested that when dissolved in toluene, aniline 
may exist to some extent in the condition of duplex molecules, the pro- 
portion of duplex to simplex molecules varying with the concentration. 
Biichner 27° discussed the limited miscibility of liquids and the 
