3:2 PROCEEDINGS OP THE AMERICAN ACADEMY 



ings, Vol. XIII. page 57, using however a porcelain tube heated by 

 a gas furnace in place of a tube of glass. The bromide of cadmium 

 when thus sublimed crystallizes in pearly scales around the open 

 mouth of the tube. 



Prepared as we have described, anhydrous bromide of cadmium is a 

 splendid preparation. The precise form of the highly lustrous crystals 

 could not be distinguished, but the scales are tabular crystals having a 

 single optical axis normal to the extended face of the scales and hav- 

 ing a high negative double refracting power. Although so soluble, the 

 crystals are not sensibly hygroscopic, and can be weighed on an open 

 watch-glass without the slightest variation of weight during the process. 

 They dissolve at once in water without leaving a trace of residue. 



The following determinations were all made with the material we 

 have described, and since only a few grammes could be sublimed at a 

 time, the separate analyses were made with the products of nearly as 

 many sublimations. Some of the material was sublimed twice, and 

 the constancy of the composition under these circumstances is the best 

 proof possible of the definiteness of the compound. The results of the 

 analyses may be classed under two heads. 



In one series of determinations, the bromide ,of antimony, whose 

 weight had been accurately determined, having been dissoh'ed in pure 

 water in a stoppered flask, a little less than the calculated amount of 

 silver required to precipitate the bromide was carefully weighed out 

 and dissolved in a measured amount of weak nitric acid allowing as 

 before described a slight excess. This solution jjroperly diluted was 

 then gradually added to the solution of bromide of cadmium under con- 

 stant agitation in order to avoid the aggregation of the bromide of 

 silver in lumps. The agitation was frequently renewed until the pre- 

 cipitate settled, and then a standard solution of silver (one gramme to 

 the litre) was cautiously added until the precipitation was complete, 

 and the last drops did not produce the faintest opalescence after stand- 

 ing. The precipitate was now washed by decantation five times suc- 

 cessively, using each time a volume of cold distilled water equal to the 

 volume of the original solution, and pouring off the water into a porce- 

 lain crucible, from which it was drawn by the reverse filter. Finally, 

 the precipitate was washed into the crucible, and after drying at from 

 120° to 130° weighed with the small filter. 



In a second series, the determinations were conducted in the same 

 way as in the first, excepting that the chief attention was directed to 

 determining the exact point of complete precipitation. In several 

 cases, indicated by an asterisk, both the amount of silver required, and 



