OP ARTS AND SCIENCES. 19 



adopted baric nitrate as its starting point. This salt, which had 

 been recrystallized ten times, was dissolved in hot water and 

 treated with the calculated amount of the purest obtainable potas- 

 sic hydrate in a platinum bottle. The resulting baric hydrate was 

 recrystallized ten times from hot water, without being removed 

 from the bottle ; but the spectroscope still showed noticeable traces 

 of potassium upon the usual fractional treatment. The hydroxide 

 was then precipitated three times successively from aqueous solu- 

 tion by means of pure alcohol, the precipitate being washed each 

 time with alcohol, with the aid of the filter pump. Even t In- 

 second, mother liquor showed no trace of potassium to the most 

 careful scrutiny. 



The pure baric hydrate thus prepared was dissolved in pure 

 water in the platinum bottle, boiled for some time to drive off the 

 alcohol, transferred to a Bohemian flask, and saturated with pure 

 bromine. This substance had been prepared as described on page 

 17, with the additional treatment of solution in pure calcic bro- 

 mide and several distillations. The mixture of baric bromide and 

 bromate was evaporated, powdered, and gradually raised to fusion 

 in a platinum vessel. The mass was gray before fusion and pale 

 green afterwards. The greenish cake was dissolved in water, 

 filtered, acidified, crystallized, dried and fused; and then this same 

 round of operations -was again repeated. The last pure white cake 

 of baric bromide was dissolved, the solution filtered, and after 

 being very faintly acidified with hydrobromic acid was again crys- 

 tallized. The final crystals were washed four times with alcohol, 

 and allowed to dry in the air. In the table below (page 27) they 

 are designated No. III. (Analyses 10, 11). 



Since baric hydroxide is so easily recrystallized it was hoped 

 that a pure preparation might be obtained directly in this way from 

 the baryta of commerce. It has been already said that five recrj B- 

 tallizations remove the chlorine; five more remove the last traces oi 

 calcium. When, however, after seventeen recrystallizations, the 

 large amount of strontium present did not seem to be considerably 

 diminished, this method was abandoned as a hopeless one. 



A long series of qualitative and quantitative experiments apon 

 the fractional precipitation of baric carbonate by the action of small 

 amounts of carbon dioxide upon baryta water showed thai this pro- 

 cess also was utterly unfitted for the complete separation of stron- 

 tium from barium, and accordingly this method was abandoned. 

 The description and data of these experiments would require much 



