Solubility of Salts in Water. 



379 



nitrates were purified in various ways ; the commercially pure 

 salt was precipitated by alcohol, or by pure nitric acid ; in 

 other cases subjected to repeated crystallization, and dried as 

 above or fused. Too much care cannot be taken to ensure the 

 use of pure salts in all saturation experiments ; for while a 

 trace of impurity may not make its presence felt in other 

 solution experiments, it is sure to vitiate the results of satu- 

 ration, inasmuch as in these cases there is always present a 

 considerable amount of undissolved salt ; and out of this the 

 whole of the impurity is dissolved, thus raising unduly the 

 percentage of salt dissolved. In this is to be found the expla- 

 nation of the serious differences so often to be observed in all 

 saturation experiments. 



The salt solutions (solvent solutions) having been thus pre- 

 pared, they were in several instances analysed by evaporation, 

 in order to ascertain whether the method of preparation was 

 sufficiently accurate for the purpose in view. The results are 

 given in Table IY. 



Table IV. 



Salt. 



Calculated. 



Found. 



Difference. 



2NaCl 



4NaCl 



6-5 

 13-0 

 19-5 



5-618 

 16-854 



6-5172 

 13-0162 

 19-520 



5-623 

 10-860 



+0-0172 

 + 0-0162 

 +0-020 

 +0005 

 +0006 



6NaCl 



lKNOy 



3KN0 3 



These determinations showed that this direct method of 

 preparing the solutions was all that could be desired, as an 

 error of 0*02, or one in 5000, is much less than that met with 

 in even the most careful saturation experiments. 



The solvent solutions were then saturated with the other 

 salts, and the results with the eight pairs of salts are given 

 in the following Tables V.-VIII. The Tables are arranged 

 in the order already given, and each consist of two parts cor- 

 responding to the two series. The arrangement in each table 

 is the same. In the first column is given the number of mole- 

 cules of the second salt per 100 molecules of water, in the 

 second the amount of salt per 100 parts of water that the above 

 corresponds to ; in the third column the salt actually found 

 (M.S. mixed salt) per 100 of water; in the fourth the mean 

 of the experimental results ; and in the fifth the difference 

 between the fourth and second columns, that is, the amount 



