GOODWIN AND HASKELL. — CONDUCTIVITY OF SOLUTIONS. 411 



In connection Avith the above data, attention is called to the close 

 agreement in the case of potassium chloride between our values of the 

 equivalent conductivity, given in column five marked "water sub- 

 tracted, " with those of Kohlrausch and Maltby ; also to the fact that 

 the application of the first correction method to this neutral salt causes 

 a change of less than 0.1 per cent in the values of the equivalent conduc- 

 tivity. It will be seen, on the other hand, that, in the case of the acids, 

 both methods of computation raise the equivalent conductivity on an 

 average by about 2 per cent in 0.25 millinormal, and by about 0.5 per 

 cent in millinormal solution. Although the values in column five vary 

 quite widely in the different series owing to differences in the quality of 

 the water used, yet when corrected by Method I the/ are in substan- 

 tial agreement in all cases where the concentration corresponding to 

 the first added portion exceeds 0.05 millinormal. The agreement 

 between the results obtained by tlie two methods is considered on 

 page 413. 



The mean of the " water-subtracted " values at the concentration mil- 

 linormal (374.0 for hydrochloric acid and 371.4 for nitric acid) are nearly 

 one per cent lower than the corresponding values earlier obtained by 

 Kohlrausch* (377 and 375 respectively). This fact led us to make a 

 new series of measurements on the conductivity of hydrochloric acid at 

 higher concentrations. These were made at 18° in a U-shaped appa- 

 ratus with a new sample of acid prepared and standardized as before.f 

 The dilutions of the stock solution (0.13747 normal) were made by 

 weight and reduced to volume-concentrations at 18°, by means of the 

 specific gravities of water and of the stock solution. On account of the 

 wide range of concentrations measured it was found necessary to use two 

 resistance vessels. The cell constants or capacities of these, determined 

 by five 0.01 normal potassium chloride solutions, were 2.5417 and 

 8.5370. The average deviation of the separate values from the mean 

 was less than 0.03 per cent in each case. 



The results are contained in the following table. Values correspond- 

 ing to the round concentrations are summarized in Section VII (page 

 415). 



The values are seen to be only slightly (0.1-0.4 per cent) lower than 



* Kohlrausch and Holborn, Leitvcrmogen der Electrolyte, p. IGO. 



t Tlie analytical data were as follows : 



Grams HCl solution taken . . . 112.34 10G.58 91.31 



Grams AgCl obtained 2.2121 2.0988 1 7994 



Equivalents per 1000 grams sol. . 0.13734 0.13734 0.13744 



