324 ON THE CALCULATION OF THE CONDUCTIVITY OF 



a concentration of 0.6 gramme-equivalent per litre. Above that 

 concentration the solutions, which were to be used in obtaining 

 data for the calculations, were made up outside the cell. Two 

 sets of solutions made in the cell in which the concentrations of 

 the final solutions were calculated by a comparison of their con- 

 ductivity with the conductivity of other solutions of approxi- 

 mately the same, and of known, concentration, were used to give 

 an idea of the course of the conductivity-concentration curves, 

 between 0.7 and 1 gramme-equivalents per litre, but in none of 

 the mixtures calculated and tabulated at the end of the paper 

 does the determination of the ionisation coefficients require the 

 use of these portions of the curves. 



To test the validity, within the limits of experimental error,, 

 of the assumption, that, in the case of dilute solutions, the 

 volumes of the latter are increased by the volume of water 

 added, I made use of Kohlrausch and Hallwachs' observations 

 of the specific gravity of solutions of sodium chloride.* For 

 solutions whose concentration is not greater than 0.2 gramme- 

 molecule per litre, the specific gravity may be represented to 1 

 in the fifth place of decimals by the equation 



s = I + . 04244? - . 003m 2 , 



where s represents the specific gravity at 18C and m the 

 concentration in gramme-molecules per litre. The following is- 

 a comparison of the values obtained by this equation with those 

 obtained by Kohlrausch and Hallwachs : 



Writing then k for the first constant in the formula, and I for 

 the second, W for the weight of water in grammes, w for the 

 weight of salt when divided by a the value of the molecular 



Wicd. Ann., 53 (1894) p. 14. 



