OP ARTS AND SCIENCES. 11 



nitrate into the sulphate is also difficult because of the well known 

 occlusion of one salt by the other. Moreover, supposing the analy- 

 sis by either method to have been satisfactorily performed, the data 

 furnished would give only the worst possible foundation for the 

 calculation of the atomic weight of barium.* Many qualitative 

 and quantitative experiments led to the complete rejection of baric 

 nitrate as the material for analysis. 



Baric bromate is very readily prepared in a pure state by a few 

 successive crystallizations, and it was hoped that this salt would 

 furnish especially valuable testimony upon the case. But investi- 

 gation showed that it was impossible to be certain that the crystal- 

 lized salt did not contain an excess of occluded water. Upon the 

 other hand, it is doubtful if all the water of crystallization can be 

 expelled without a slight decomposition of the salt. Since water is 

 the one impurity most to be dreaded in all such work, baric bromate 

 was rejected, except as a means of obtaining the bromide in a pure 

 state. 



The carbonate was next experimented upon; and while the re- 

 sults were more promising than those from the nitrate and bromate, 

 they were less satisfactory and conclusive than those obtained from 

 baric bromide. 



The advantages of the use of a bromide for an investigation upon 

 atomic weights are so manifest, and have been so often discussed, as 

 to need no further mention. The current descriptions of the de- 

 liquescence and instability of the baric salt alone postponed the 

 consideration of this substance. Investigation showed that mis- 

 leading statements about the salt have found their way into chemical 

 literature. In reality, the substance is as well adapted for accurate 

 work as baric chloride and most other materials upon which we 

 must rely. 



The Properties of Baric Bromide. 



Baric bromide crystallizes from aqueous or dilute alcoholic solu- 

 tions in doubly terminated monoclinicf prisms, which are somewhat 

 hygroscopic, but not deliquescent in ordinary weather. 



The cr3-stallized salt contains two molecules of water, together 

 with the slight excess which is usually to be found in such crystals. 



* See Ostwald, Allgemein. Chem., I. 23. 



t Werther, Journal fiir prakt. Chem., XCI. 167. Also Von Hauer, same 

 Journal, LXXX. 230. Rammelsberg states that the salt is isomorphous with 

 baric chloride (Pogg. Ann., LV. 237). 



