54 president's address — section' b. 



I propose to-day very briefly to put before you some of the 

 results of investigations on the solvent action of inorganic sub- 

 stances other than water, and to point out as far as may be possible 

 the similarities and dissimilarities in their general behaviour. 



Of the large number of solvents that have been investigated, 

 mention may be made of the following : — Water, alcohols, acetone 

 and ether, the halogen hydrides, sulphuretted hydrogen, ammonia 

 and certain organic amines, hydrogen cyanide and the alkyl 

 cyanides, bromine, sulphur dioxide, nitrogen peroxide, nitric acid, 

 some of the trihalides and oxyhalides of phosphorus, arsenic and 

 antimony, etc. etc. 



Exhaustive investigations have been carried out with only a 

 few of these, and I propose to discuss briefly the behaviour of solu- 

 tions in water, ammonia, sulphur dioxide, and the halogen hydrides. 



Solutions in any solvent may be classified as conducting and 

 non-conducting, according as they do or do not permit of the easy 

 passage of a current of electricity. A corresponding classification 

 of solutes into electrolytes and non-electrolytes has been made 

 and based again solely on the results of the study of aqueous 

 solutions. To the former class belong the acids, bases and salts, 

 and to the latter substances such as sugar, the alcohols, ethers and 

 other non-saline solutes. 



It is worth pointing out that no metallic salt is known which 

 'dissolves in sulphuretted hydrogen or in any of the halogen 

 hydrides, and that in all of these solvents substances hke ether, 

 alcohols, ketones, etc., dissolve more or less freely with the forma- 

 tion of conducting solutions, and might therefore have been regarded 

 as electrolytes had these solutions been studied before those in 

 water. 



The selective action of solvent towards solute in forming a 

 conducting solution is a fact which speedily becomes apparent 

 when a number of solvents are considered. Take, for example, tlie 

 following figures, which represent the conductivities of equivalent 

 solutions of potassium iodide and of hydrogen chloride in the same 

 series of solvents : — 



solvent. 



SOLUTE. Methyl Sulphur 



Water. Alcohol. Acetone. Dioxide. 



Potassium Iodide . . 143 98 154 45 



Hydrogen Chloride . . 360 133 2.21 extremely small. 



The lack of any regularity in these figures is very striking, and 

 the consideration of a larger number of instances would convince 

 us that it would be more correct to speak of conducting solutions 

 than of conducting solvents, as is so frequently done. 



That attention has in the past been devoted mainly to the 

 solvent is evidenced by the large amoimt of work which has been 



