1902.] on the Ions of Electrolysis. 35 



mixtures which at ordinary temperature conduct well, the supposi- 

 tion is not unnatural that it is mixture that makes electrolytes good 

 conductors. And again, if what has been said is correct, we must, in 

 order to have good conduction, protect the wandering constituents 

 from frequent meeting with one another, and this service is performed 

 by the solvent, which makes it possible for the ions to get over a 

 part of their journey — and so much larger a part the more solvent 

 there is — without re-forming molecules. It is this suggestion which 

 I ventured a little while ago to put into an allegorical form. 



In order to compare the conductivity of one electrolyte with that 

 of another, it is necessary that we choose comparable quantities of 

 the two, and there is no difficulty in seeing that such comparable 

 quantities are those decomposed by the same current of electricity — 

 that is to say, the electro-chemical equivalents of the electrolytes. 

 Accordingly, instead of expressing the concentration in percentage 

 of the solute, Kohlrausch uses " molecular numbers." The molecular 

 number of a solution is the quantity, in grams, of the solute con- 

 tained in a litre of the solution divided by the equivalent of the 

 solute. Dividing the conductivity of a solution by its molecular 

 number gives its molecular conductivity. It will be seen that 

 " molecular " is not used here in its ordinary chemical sense, but as 

 the meaning is quite distinctly stated no confusion need arise. 

 Kohlrausch showed that the molecular conductivity increases as the 

 solution becomes more dilute, and with extreme dilution approaches 

 a constant value. 



I now show an experiment to illustrate this. 



The apparatus * consists of an electrolytic cell in the form of a tall 

 rectangular trough, the back and front being broad plates of glass, 

 while the sides are composed of narrow strips of wood completely 

 lined with silver foil. The bottom of the cell is made of non-con- 

 ducting material. The two sheets of silver serve as electrodes, being 

 connected to binding screws by means of external wires. The cell 

 is introduced into a battery circuit along with a galvanometer of low 

 resistance. If the cell be filled with pure water there is scarcely an 

 appreciable current transmitted. On removing the water and pour- 

 ing in 20 cc. of a d-normal silver nitrate solution, so as to cover the 

 bottom to a depth of a few millimetres, a current passes as indicated 

 by the galvanometer. If pure water be now added in successive 

 portions and the solution stirred after each addition, an increase in 

 the strength of the current is observed, the increase being greatest 

 after the first dilution, and becoming less with each succeeding dilu- 

 tion, so that a maximum is approached. In this experiment the 

 distance between the electrodes is constant, and the area of the 

 electrodes and of the cross-section of the conducting solution are 

 proportional to the volume of the solution, and the quantity of the 



♦ From a paper by Noyes and Blanchard, in the Zeitschrift fiir physikaliBche 

 Chemie, xxxvi. p. 9 (1901). 



D 2 



