512 ALUMINUM. 



must again be carefully washed, as before, to get rid of every 

 portion of the liquid in which it was precipitated. By drying 

 in air, alumina is reduced to a few hundredths of the bulk of 

 the humid mass. It is still a hydrate, but when ignited at 

 a high temperature, it gives pure alumina. One hundred parts 

 of alum furnish only 10.3 parts of alumina. 



Alumina is white and friable. It has no taste, but adheres to 

 the tongue. Before the oxihydrogen blow-pipe, it melts into a 

 colourless glass. After being ignited, it is dissolved by acids 

 with great difficulty. It is highly hygrometric, condensing 

 about 15 per cent of moisture from the atmosphere in damp 

 weather. If ignited alumina contains a small portion ot 

 magnesia, it becomes warm when moistened with water ; this 

 property is very sensible, even when the proportion of magne- 

 sia does not exceed half a per cent. It appears to be due to 

 heat disengaged by humectation, a phenomenon first observed 

 by Pouillet. 



The hydrate of alumina, when moist, is gelatinous and semi- 

 transparent, like starch, but dries up into gummy masses. It is 

 completely insoluble in water, but is readily dissolved by acids, 

 and also by the fixed alkalies. Caustic ammonia dissolves it, 

 only in very small quantity. When an excess of the hydrate, 

 immediately after precipitation, is digested in caustic potash, 

 by a moderate temperature, and the solution filtered and sealed 

 up in a flask, there separate from it after a time crystals upon 

 the sides of the vessel. The same crystals are obtained, on 

 allowing the solution to absorb carbonic acid slowly from the 

 air. They are white and transparent at the edges, and contain 

 34.61 per cent of water, or 3 equivalents, which they do not 

 lose at 212. (Mitscherlich's Traite.) The mineral gibbsite is a 

 native hydrate of alumina of the same composition, A1 2 O 3 + 

 3HO. Another native hydrate exists, containing less water, 

 A1 2 O 3 + 2HO. It is called diaspor by mineralogists, from de- 

 crepitating and falling to powder when heated, a property which 

 the artificial hydrate in gummy masses likewise exhibits. 



Hydrated alumina has a peculiar attraction for organic matter, 

 which it withdraws from solution, and hence this earth is apt 

 to be discoloured when washed with water not absolutely pure. 

 This affinity is so strong, that when digested in solutions of 

 vegetable colouring matters^ alumina combines with, and carries 

 down the colouring matter, which is removed entirely from the 



