386 Mr. T. 0. Fitzpatrick on the Action of 



Water Solutions. — The conductivity-values for equivalent 

 solutions of the different salts are all of the same order, with 

 the exception of the mercuric-chloride solutions, the peculiar 

 behaviour of which has already been alluded to. 



The chlorides are better conductors than the corresponding 

 nitrates, though not to any great extent. The behaviour of 

 magnesium chloride is anomalous, the conductivity-values for 

 its solutions being almost exactly one half those of the cor- 

 responding calcium-chloride solutions. The same is not the 

 case for the nitrate, though its conductivity is less than that 

 of the calcium nitrate ; a fresh solution of the chloride gave 

 the same values as the former ; the solution was carefully 

 tested for calcium or ammonium chloride, but not a trace was 

 found. 



It would appear, therefore, that a solution of magnesium 

 chloride containing 1 equivalent has the same conductivity 

 as one of calcium chloride containing half an equivalent. 



For the ferric-chloride solutions the conductivity is too 

 high for the more dilute solution, the conductivity not being 

 proportional to the quantity of salt in solution ; this is pro- 

 bably due to the decomposition of the ferric into ferrous 

 chloride. 



Alcohol Solutions. — For none of the alcoholic solutions is 

 the conductivity proportional to the amount of salt in solution ; 

 the conductivity not diminishing directly with the increase of 

 the dilution. 



Further, the conductivities vary considerably for the 

 different salts. 



The lithium salts have values comparable with one another, 

 and the values for the calcium and magnesium salts are of 

 the same order ; but the values for the lithium salts are 

 between 10 and 20 times as great as the conductivities for 

 the corresponding solutions of the other salts. 



The same is true of the methyl-alcohol solutions, though 

 the lithium salts conduct only about twice as well as the 

 calcium salts in methyl-alcohol solutions. 



Next, comparing the conductivities of equivalent solutions 

 for the same salt and the different solvents, it is clearly seen 

 how largely the values for the conductivities of saline solu- 

 tions depend on the character of the solvent. 



Prof. Arrhenius considers that in dilute aqueous saline 

 solutions, the action of the water is to cause the dissociation of 

 the salt ; but how this action is exerted he does not suggest. 



Other solvents do not exert such dissociating action, he 

 continues, and hence do not conduct. This is not quite exact, 

 as, though the alcoholic solutions do not conduct as well as 



