ON THE HALOGEN HYDRIDES AS CONDUCTING SOLVENTS. 149 



associated, all four compounds are equally able to act as conducting solvents, whilst 

 this property is not possessed by hydrogen phosphide, which is associated. 



BRUHL (' Zeit. Phys. Chem.,' 1898, 27, p. 319) has pointed out that unsaturated 

 compounds, as a rule, are good conducting solvents. A consideration of the non- 

 conducting unsaturated solvents phosphorus hydride and trichloride, and of the 

 conducting saturated solvent phosphorus oxychloride, is sufficient to show that this is 

 not a general rule. 



The heat of vaporisation (OBACH, 'Phil. Mag.,' 1891, (5), 32, p. 113) is a fourth 

 property which has been suggested as being intimately connected with the dissociating 

 power of the solvent. In this case, as in that of the others considered, the connection 

 is very obscure and many exceptions occur. 



The temperature coefficients of conductivity and of viscosity are approximately 

 equal in the case of aqueous solutions. This is not so in solutions in the solvents 

 examined by us, although in these, also, a rise of temperature conditions an increase 

 of conductivity and a decrease of viscosity. It is interesting to note that the 

 increase of conductivity is, in nearly all cases, greater than the decrease of 

 viscosity. 



The foregoing summary shows that failure has attended every attempt which has 

 been made to express the power of forming conducting solutions as a function of the 

 solvent only. 



As a matter of fact, every solvent exhibits a very marked selective action an 

 regards the nature of the conducting solute. Thus water dissolves the majority of 

 salts to form solutions which conduct the current ; organic bodies also, other than 

 salts, are in many cases soluble, but the solutions are not conductors. Hydrocyanic 

 acid behaves similarly to water, but only a few salts are appreciably soluble in this 

 solvent. Ammonia, sulphur dioxide, and some other solvents form conducting solu- 

 tions, not only with many salts, but also with a few organic substances not usually 

 classed as electrolytes. The halogen hydrides, on the other hand, form conducting 

 solutions with non-saline organic substances, as well as with salts of the ammonium 

 bases, but such solutions are not formed with metallic salts. 



It is evident, therefore, that the ability to form a conducting solution is a function 

 of both the solute and the solvent, and this has been recognised in the various 

 attempts that have been made to connect the ionising power of a solvent with its 

 tendency to form compounds with the solute. Indeed CADY ('Jour. Phys. Chem.,' 

 1897, 1, p. 707) was led to investigate the conductivity of solutions of substances in 

 ammonia from the analogy between the water and the ammonia compounds of copper 

 sulphate. 



KAHLENBERG and SCHLUNDT (' Jour. Phys. Chem.,' 1902, 6, p. 447) express the 

 opinion that conductivity is due to mutual action between the solute and the solvent ; 

 and an attempt to obtain experimental evidence in support of this view has been 

 made by PATTEN (' Jour. Phys. Chem.,' 1902, 6, p. 554). 



