Mr. Crum on the Acetates of Alumina. 309 



When alum, dissolved in six times its weight of water, was treated 

 with three equivalents of acetate of lead — enough to decompose its 

 tersulphate of alumina, but not the sulphate of potash, a large propor- 

 tion (74 per cent.) of the sulphate of potash was found to be carried 

 down by the sulphate of lead formed in the process. To a solution so 

 prepared, the loss was restored by an addition of sulphate of potash, and 

 a mixture formed of — 



KO, SO3 together with AI2O3, 3A 

 This solution, diluted so as to contain about 0'3 per cent, of alumina, 

 (from 5 oz. of alum per gallon,) was exposed to heat. At about 90" Fahr., 

 the gelatinous precipitate began to form ; and after two hours boiling, when 

 scarcely a trace of alumina remained in solution, it was thrown upon a 

 filter of calico, which, was kept hot within a steam pan until the filtrate 

 passed through. The collected precipitate was pressed between numerous 

 folds of bleached calico, and brought gradually to the state of a stiif clay 

 occupying not more than ^-^ of the volume of the original solution. It 

 was then divided into portions which could conveniently be shaken 

 and thoroughly mixed in bottles with quantities of water equal in all to 

 the original solution. The mixture was filtered as before, at nearly a 

 boiling heat, and pressed between folds of calico ; and another repetition 

 having been made of the same process, the precipitate was deprived of 

 every admixture of soluble matter. 



A portion being dissolved in nitric acid and filtered, gave an abun- 

 dant precipitate with nitrate of baryta. Exposed to a white 

 heat, and the residue pulverized and boiled in water, the filtered 

 liquor gave no longer a precipitate with nitrate of baryta, and no 

 indication, or a very doubtful one, of alkali to litmus paper ; proving 

 that the sulphuric acid indicated by the first test had been combined, 

 not with potash, but with alumina. Mixed with sulphuric acid the 

 substance gave no smell of acetic acid, and when burned with oxide of 

 copper it yielded no more carbonic acid than could readily be accounted 

 for by the few fibres of cotton wool that were to be distinguished in the 

 liquid. The substance dried into a hard, whitish, semi-transparent 

 matter, easily pulverized. 



For analysis 33-03 grains dried at 100° Fahr. were dissolved in muriatic 

 acid, and the solution was exactly decomposed at a boiling heat by 15-25 

 grains of chloride of barium.* This indicates 5-867 grains, or 17-29 per 



* The heat in this case greatly facilitates the arrival at the nearest point of 

 decomposition, particularly by enabling the filter to furnish, immediately, clear 

 portions, from time to time, for testing. 



Where nitrate of baryta is emj)loyed for decomposing a sulphate, it was 

 observed by Mitscherlich that a quantity of the nitrate was taken down unaltered 

 by the sulphate of baryt.u This source of error, which amounts to not less than 

 5 per cent, of the ijuantity of sulphuric acid when the experiment is performed in 

 the cold, is entirely obviated at a boiling heat. 



Boiling, however, docs not render the decomposition of a sulphuric jalt by a 



