102 PROCEEDINGS OF THE AMERICAN ACADEMY. 



referred to water at the same temperature, the specific gravity of the 

 bromide must be 4.64. Nickelous bromide is iusoluble iu toluol. 



Purification of Materials. 



Nickel. — Since the jjossibility of preparing and analyzing pure nick- 

 elous bromide had -now been proved, the next step was to purify all the 

 materials concerned in its manufacture. First among these materials 

 comes nickel, which must not only be freed from all known impurities, 

 but must also be so treated as to detect and eliminate unknown ones. 

 With this latter purpose in view, our material was obtained from two 

 distinct sources ; first, the " pure " nickel of commerce ; and, secondly, 

 really pure nickel (containing only a little iron) jjrepared by Dr. Mond 

 through the carbonic oxide process and kindly presented by Dr. Wolcott 

 Gibbs. 



It is convenient to consider first our treatment of the commercial 

 nickel. In proceeding with the further purification of this sample, our 

 first step was to remove the metals of the copper and tin groups. The 

 simple treatment with hydric sulphide has generally been considered suf- 

 ficient to insure the separation of the metals of these groups, in spite of 

 the fact that many of the sulphides when present in small quantities 

 often assume a colloidal condition in which they cannot be separated by 

 filtration. In our case this difficulty was avoided by regulating the 

 acidity of the solution so that a certain amount of black nickelous sulphide 

 was precipitated, which effectually "swept" this liquid, coagulating small 

 quantities of foreign sulphides. After filtration the liquid was boiled to 

 drive off" the hydric sulphide, oxidized with a few drops of nitric acid, 

 made alkaline with ammonia, and filtered. The precipitation of the 

 sulphide was now continued by the passage of a little washed hydric sul- 

 phide. This first comparatively small amount of sulphide was filtered 

 out and discarded. The remaining nickel was then as nearly as possible 

 completely precipitated in the form of sulphide. After the precipitate 

 had been allowed to settle in a closed flask over night, the liquid was de- 

 canted, and the precipitate was washed by decantation many times with 

 boiling water until neuti'al. Cold dilute hydrochloric acid was now 

 added, and the precipitate was digested for several days. Since the 

 color of the supernatant liquid showed that a slight amount of even 

 the comparatively insoluble nickel sulphide had gone into solution, the 

 assumption was not unreasonable that nearly, if not quite, all of the more 

 soluble sulphides must have been removed. The black precipitate was 

 now repeatedly washed with hot water until the washings were quite 



