294 Van Name and Bosworth — Mixed Crystals of Silver 



exactly the same strength by comparison with the same sodium 

 hydroxide solution. Silver sulphate was prepared from the 

 purchased salt by dissolving in sulphuric acid, precipitating by 

 diluting with water, and washing on the "filter until the filtrate 

 showed no acidity. Silver chromate was prepared by evaporat- 

 ing an ammoniacal solution of the ordinary precipitated pro- 

 duct, a method which gives the salt in a finely crystalline state, 

 in which it is very easily washed. 



The method in detail was as follows : Two liters of the 

 standard sulphuric acid solution were warmed to about 50° 

 and the desired amounts of silver sulphate and chromate dis- 

 solved as described above. This solution was filtered while 

 still warm into a large beaker, cooled to nearly 25°, and placed 

 in the thermostat. A rotary stirrer kept the liquid in con- 

 stant motion during its crystallization. After 24 hours, sam- 

 ples of the liquid were taken for analysis, and the crystals 

 collected, washed, and dried at 130°. The crystals were small, 

 and w T hen examined under the microscope were found to be well 

 developed and apparently free from inclusions of liquid. 



In analyzing the solution Ag, S0 4 , and Cr 2 7 were deter- 

 mined in separate portions of 25 ce each, the first two by weigh- 

 ing as AgCl and BaS0 4 respectively, the last volumetrically 

 by titrating the iodine set free from KI. The same methods 

 were used in analyzing the crystals. The results are given in 

 the accompanying table. 



Owing to the small differences in composition between the 

 different solutions analyzed, and to the high ratio of S0 4 to 

 Cr 2 7 , experimental errors have a very disturbing effect, as is 

 evident on comparing the values obtained, given in the first 

 four columns of the table.* The next four columns contain 

 the molecular ratios of S0 4 to Cr 2 7 in the solution and crystals 

 respectively. The ninth column gives the total acidity in 

 equivalents per liter, calculated by subtracting the equivalent 

 concentration of the silver from the sum of those of S0 4 and 

 Cr 2 7 . The observed slight variations in the total acidity 

 were probably caused in part by difference in the amount of 

 evaporation which took place during the experiment, and in 

 part by loss of chromic acid through the crystallization of 

 dichromate. 



In experiment I no chromate was present. Experiments II 

 and III gave mixed crystals containing different proportions 

 of dichromate and of different color, vermillion in II and deep 



* Thus in experiment IV the proportion of dichromate in the solution 

 must in reality have been larger than in experiment II, although the analy- 

 sis, probably owing to a summation of errors, indicated the reverse. The 

 large effect of certain errors is well illustrated by the fact that a difference 

 of less than 0'2 CC in the amount of thiosulphate used in experiment IV 

 would have eliminated this inconsistencv. 



