354 ON THE CALCULATION OF THE CONDUCTIVITY OF 
constituent simple solutions of the equi-molecular mixtures, as 
the case may be; (2) their concentrations of ions; (3) the 
dilutions, in litres per gramme-equivalent, of the respective salts 
in the regions of the solutions occupied by them, determined by 
the graphical process referred to above, on the hypothesis that 
the salts exist separately, (4) the values of the conductivities as 
measured, and (5) as calculated, and (6) the excess per cent. of 
the former values over the latter. The concentrations of ions, 
that is the ionization coefficients divided by the dilutions, and 
the conductivities, are given in terms of the units before 
specified. The solutions, the values of whose concentrations are 
printed in italics, were made up in the cell, 
In the case of the double salt solutions from I gr.-eq. per litre 
down to 0.4, the values of the conductivities are less than they 
would be if the salts were separated, as judged from the values 
calculated on that hypothesis. For this range of solutions the 
signs in the Jast column are all positive, and the differences are 
beyond the limits of error, though ia two cases not much beyond. 
With the first four solutions the excess of the calculated values 
does not differ much, though the fourth is unexpectedly high. 
On the first addition of water to the 0.5019 solution, the difference 
becomes much smaller and continues so as water is added. 
Moreover, with most of the solutions from 0.5 to 0.37, that is, 
with the solutions which were diluted in the cell, the conduc- 
tivity rose after the water had been added and mixed. In these 
cases the last value observed was taken as the conductivity of 
the solution. Thus the first measurement of the conductivity of 
the 04075 solution gives a value 1 per cent. Jess than the calcu- 
lated value; but the last two measurements of the same solution 
taken half an hour later give a value only 0.6 per cent. less. In 
the case of the 0.4183 solution, however, no change was noticed. 
The effect is as if the addition of wacer, and possibly also the 
stirring of the solution, caused the double salt to undergo rapid 
dissociation. But it may have been due to the thorough mixture 
of the original solution and the water added, requiring time. 
In the case of the dilute solutions the differences are within 
limits of error, althongh their signs are not entirely satisfactory 
