RICHARDS AND CUSHMAN. ATOMIC WEIGHT OP NICKEL. 109 



excess of argentic nitrate, and the argentic bromide was collected and 

 weighed with the usual precautions. 



When the manipulation had tlius been mastered, the art of preparing 

 absolutely pure nickelous bromide had been perfected (see Analysis 7), 

 and the atomic weight of nickel had been approximately determined, the 

 method of procedure in subsequent analyses was changed. The now 

 perfectly clear solution was treated with just enough argentic nitrate, 

 prepared from the purest weighed silver, to complete the precipitation. 

 The mean between the two possible end points was determined by titrat- 

 ing backwards and forwards with hundredth normal argentic nitrate and 

 hydrobromic acid solutions ; * and thus was determined the ratio of silver 

 to nickelous bromide entitled Series III. After this end point had been 

 determined, a slight excess of argentic nitrate was added to the solution, 

 and the whole was violently shaken. The precipitate was collected upon 

 a Gooch crucible, washed with water containing a trace of argentic ni- 

 trate, later with pure water, and finally dried and weighed. The traces 

 of asbestos carried away by the wash water were of course determined, 

 and all the usual precautions were taken to insure great accuracy. 

 Thus was obtained the series of results given in Series II. 



THE ATOMIC WEIGHT OF NICKEL. 

 = 16.000; Ag= 107.93. 

 Fist Series (Preliminary). Ratio = 2 AgBr : NiBr2. 



* For the details see These Proceedings, XXX. 384. 



