1018 
PROFESSOR J. W. MALLET OR A REVISION 
last bright yellow heat in a gas furnace of Fletcher's construction. This temperature 
was maintained for a full hour. On cooling down, the smaller crucible was taken out, 
the cover cautiously raised, and enough of a strong solution of ammonium carbonate 
introduced to moisten the alumina. Drying gently in a steam-bath, the crucible was 
re-ignited over an alcohol blast lamp, producing a strong red heat, and the addition 
of ammonium carbonate, drying, and ignition once repeated. As soon now as the 
crucible had ceased to be visibly red hot it was placed in the desiccator as at first, 
allowed to cool down to the temperature of the balance-room, quickly transferred to 
the weighing bottle, the stopper of which was inserted, and the final weighing was 
made while the alumina in the crucible was thus protected from absorption of moisture 
from the air. These experiments were carried out during a period of remarkably 
steady weather, with very little variation of atmospheric temperature, pressure, or 
moisture in the balance-room during the whole series. 
Direct results of first series of experiments .—The results were as follows :— 
A.—Alum dried by exposure to air for 2 hours at 21°-25° C. 
I.— 8’2144 grins, of (NH 4 ,) :3 Al 2 (S0 4 ) i ,.24H 3 0 left 
II.—14-0378 
TIL— 5-6201 
IV.—11-2227 
V.—10-8435 
9258 grm. of Ah0 3 . 
5825 
6337 
2657 
2216 
B.—Alum dried by exposure to air for 24 hours at 19°-26° C. 
VI.—12-1023 grins, of (NH 4 ,)oAb(S0 4 ,) 4 ,.24H 2 0 left 1"3660 grm. of ALCh 
VII.—10-4544 
VIII.— 6-7962 
IX.— 8-5601 
X.— 4-8992 
33 
3 3 
33 
33 
1-1796 
•7670 
"9654 
•5528 
33 
33 
Second series of experiments. 
Preparation and purification of aluminum bromide. —Aluminum bromide was pre¬ 
pared by the action of liquid bromine upon metallic aluminum of commerce, and was 
afterwards carefully purified. The first action is so violent that without special pre¬ 
caution the process involves some danger. In a first attempt a lump of aluminum 
weighing 15 to 20 grins, was dropped into a long-necked flask containing a considerable 
quantity of bromine. There was little action for a few moments, but as soon as it 
began vivid combustion took place, torrents of bromine vapour were driven forth, and 
after the flask had cooled the surplus metal was found to have been completely fused 
and had nearly melted its way through the glass. 
