BAXTER. — REVISION OP THE ATOMIC WEIGHT OF IRON. 251 



quantity of pure silver was dissolved in pure nitric acid and the solution, 

 after dilution, was added slowly to the bromide. The resulting silver 

 bromide was filtered upon a Gooch crucible, heated to 200° in an electric 

 oven, and weighed. The asbestos shreds which passed through the cru- 

 cible, together with any accompanying silver bromide, were collected 

 upon a small filter. This filter was then ignited, the ash treated with a 

 drop of a mixture of pure nitric and hydrobromic acids, and finally the 

 weight of the residue determined. After the weight of the silver bro- 

 mide had been found, it was fused in a porcelain crucible and the loss in 

 weight determined. In Analyses 5 and 6 exactly the calculated amount 

 of silver was used in the precipitation, and portions of the clear solution 

 above the precipitate were tested with solutions of hydrobromic acid and 

 silver nitrate in a nephelometer kindly loaned for the purpose by 

 Mr. R. C. Wells. In both analyses a very slight opalescence was pro- 

 duced to the same extent in both portions, showing that the exact 

 amount of silver necessary to combine with the bromine had been added. 

 The absence of any appreciable amount of precipitate in either portion 

 is evidence that no chlorine was present in the salt; for silver chloride, 

 owing to its greater solubility, would have made itself evident by a con- 

 siderable cloudiness in the nephelometer test. In these two analyses an 

 excess of silver nitrate was then added and the weight of the silver 

 bromide was determined (Analyses 3 and 4). 



In order that the weight of the ferrous bromide might be reduced to a 

 vacuum standard, its specific gravity was found by determining the 

 weight of kerosene displaced by a known quantity of salt. The kerosene 

 was first dried by means of stick soda, and then distilled. The portion 

 distilling between 200° and 240° was used in the determinations. Its 

 specific gravity at 25° referred to water at 4° was found to be 0.7C93. 



