LEWIS AND WHEELER. — CONDUCTIVITY OP SOLUTIONS. 429 



From this point of view we see that the increase of molecular conduc- 

 tivity with the concentration, which we have found in the case of potas- 

 sium iodide in iodine and which others have found in the case of salt 

 solutions in many organic solvents, is a deviation from the mass law in 

 greater degree, hut of the same kind and in the same direction, as that 

 which is observed in the case of aqueous solutions of strong electrolytes. 



What is the cause of this general deviation from the mass law, and 

 why should it be greater in the case of weak dissociating agents ? 



Before attempting to answer this question let us give definition to an 

 idea which we shall find useful. If we examine a number of liquids 

 chosen at random, we find that as a rule those liquids which, when pure, 

 are the best conductors of electricity, are, as solvents, the best dissociat- 

 ing agents, and when dissolved in other solvents are themselves the most 

 dissociated. In other words the property of dissociating other substances 

 and the property of being dissociated when pure and when dissolved in 

 other solvents go usually hand in hand. Thus pure nitric acid is a con- 

 ductor, it dissociates other substances dissolved in it, and is dissociated 

 in other solvents. Benzol has none of these three properties. Let us 

 call a substance possessing these properties to a marked degree an elec- 

 trophile. The best electrophiles are substances which consist of two 

 parts, one having a strong positive, and the other a strong negative 

 electro-affinity, as this term is used by Abegg and Bodlander.* In this 

 category are included most inorganic salts. Ether and the hydrocarbons 

 are types of the poorest electrophiles. 



Water occupies an intermediate position. That it stands first or 

 nearly first among known dissociating solvents is due doubtless to the 

 fact that none of the best electrophiles are liquid at ordinary tempera- 

 tures. The only liquids which are good conductors in a pure state are 

 melted salts.f If these salts could be supercooled to ordinary tem- 

 peratures their conductivity would still be far greater than that of water, 

 and doubtless their dissociating power too. 



* Zeit. anorg. Chem ., 20, 453 (1899). 



t Whether the good conductivity of melted salts is due to a high degree of dis- 

 sociation or to great mobility of the ions has not hitherto been demonstrated. One 

 of the authors together witli Mr. B. S. Lacy has attempted to answer this im- 

 portant question by measuring directly the ionic velocities in melted salts. The 

 experiments, although not yet concluded, have progressed far enough to show that 

 the mobilities of the ions are probably even less than in water. The degree of 

 dissociation must therefore be enormously greater than in the case of any other 

 pure liquids. 



