100 PROCEEDINGS OF THE AMERICAN ACADEMY 



With the same view, to anothei' solution a very dilute hydrochloric 

 acid of known strength was added until a precipitate of alumina be- 

 came visible. But the quantity of acid which could be so added 

 corresponded to only the half of one per cent of baryta ; deducting this 

 amount, the liquid contained 



BaO 61.44 



AU O3 38.56 



100 



A solution giving the same reactions may be obtained by fusing car- 

 bonate of bai'yta and alumina with other chlorides, or with salts of a 

 similar chemical nature. By employing a salt easily soluble in alcohol, 

 I hoped to be able to separate the aluminate. To this end 5 parts of 

 carbonate of baryta and 2 of alumina were fused with 10 parts of bro- 

 mide of sodium, the mass dissolved in water, and evaporated in a water- 

 bath under a pressure less than that of the atmosphere, and finally in 

 vacuo over sulphuric acid. The residue was digested with alcohol, 

 brought upon a filter and washed with alcohol till the washings contained 

 only a trace of bromine. The mass was then only partially soluble in 

 water, the solution gave the reactions of aluminate of baryta, but con- 

 tained a great deal of bromine. In fact, under these circumstances 

 a double salt of aluminate and bromide is formed which is insoluble 

 in alcohol. If the simple aluminate be really soluble in alcohol, as 

 would be inferred from the statement of M. Deville, it is not easy 

 to understand how its combination with bromide of sodium should be 

 nearly insoluble in that liquid. 



Finally, carbonate of baryta was fused with fluor-spar and alumina. 

 The fusion is easily eflfected, the fluor-spar is apparently not decom- 

 posed, and the aluminate is formed as in the former cases. The fused 

 mass was very finely pulverized and digested with water, which could 

 here dissolve nothing but the aluminate. The solution contained 



BaO 70.5 



Al, O, 29.5 



100 



which is not very far from the composition of the dialuminate already 

 described. 



Another solution made in the same way was evaporated until the 

 salt began to separate, and then precipitated with twice its volume 

 of alcohol and the precipitate washed with alcohol : it was found to be 

 composed of 



