THEORIES OF ELECTROLYSIS. 311 



It seems, however, that these two conclusions need not 

 be mutually inconsistent. Since it is evident that one salt 

 molecule can influence a large number of water molecules, 

 it follows that the chemical forces are very far-reaching. 

 Each particle of salt extends its influence over a consider- 

 able region round it, and the properties of the solution will 

 depend on its percentage composition. It is natural, then, 

 that, when the composition of the liquid passes through the 

 value which corresponds to a definite molecular compound, 

 the fact should appear in some modification of the physical 

 properties, although no definite aggregate moving about as 

 an independent physical molecule need be formed. There 

 is, therefore, no objection to the view that the ions exist 

 free enough from each other to enable them to play the 

 part assigned them by the dissociation theory. 



It is, nevertheless, remarkable that bodies, so stable 

 when solid, should be dissociated in solution, but a study of 

 the properties of the solvent throws some light on the dif- 

 ficulty. If the forces between the parts of the salt mole- 

 cule are electrical in their nature, as is quite probable, it 

 follows that, as J. J. Thomson has pointed out, 1 immersion 

 in a medium like water, of high dielectric constant, will 

 cause these forces to become much weaker. The effect is 

 best explained by imagining a large mass of conducting 

 material to be placed near two small spheres charged with 

 opposite kinds of electricity. Each charged sphere attracts 

 electricity of the opposite kind to the neighbouring parts of 

 the conducting mass, and to this induced charge most of the 

 sphere's lines of force will then converge. Its influence on 

 other outside objects will be, therefore, much lessened, and 

 consequently the force of attraction between the two little 

 spheres will be much reduced. The effect of an insulator 

 of high dielectric constant is similar in kind though rather 

 less in magnitude. This may explain the differences 

 observed in the molecular conductivities of the same salt 

 dissolved in different solvents, such as water and alcohol, 

 for example, for, other conditions being the same, the effect 



1 Phil. Mag., xxxvi., p. 320, 1893. 



