310 ON THE CALCULATION OF THE CONDUCTIVITY OF 
More dilute solutions were prepared from this one by adding 
water, and their concentrations calculated. Check analyses, 
however, were made after any portion had gone through a 
nuinber of dilutions, and, if found necessary, the caleulated con- 
centrations were corrected from these results. 
As the method of calculation required a knowledge of any 
appreciable change of volume which would occur on mixing 
simple solutions of each of the salts. of such strength as to form 
a solution of the same concentration as the solution of the double 
salt under investigation, density determinations were made’ of 
a number of such solutions, before and after mixing. These 
measurements were carried out with Ostwald’s form of Sprengel’s 
Pyknometer. They might be in error by about 5 in the fifth 
decimal place. No change of volume was found to occur on 
mixing the most concentrated solutions examined which would 
appreciably effect the calculation of the conductivity. The 
density of a mixture of the constituent solutions of the double 
sulphate was found in the case of some of the stronger solutions 
(the only ones tested) to be the same, within the limits of experi- 
mental error, as the density of a solution of the double salts of 
the same concentration. 
For the simple solutions the ionization coefficient was taken 
to be equal to the ratio of the specific molecular conductivity to 
the specific molecular conductivity at infinite dilution. The 
values of the molecular conductivity at infinite dilution used in 
the calculations were :—1280 x 1078, and 1100x10-° for Potas- 
sium and Copper Sulphate respectively, as determined by 
Kohlrausch.* I was not aware, at the time the calculations were 
made, that he had given 1270 10°, and 1120 x10 °, as better 
values for these salts.+ I have, however, repeated some of the 
calculations and find that the difference caused by using these 
later values are in all cases less than 0.06 per cent. 
* Wied. Ann., Vol. 26. p. 204. 
t Wied. Ann., Vol. 50 (1893), p. 406. 
