ARTIFICIAL PRODUCTION. 75 
ignited. Barium peroxide is also added to the top of the mixed 
oxides and ahiminum, and then a little more of the powdered 
aluminum is placed on top of the barium peroxide. The barium 
peroxide is readily ignited by a wax tai)er or other fuse, and the 
heat generated, together with the excess of oxygen, is suflicient to 
ignite the powdered aluminum. This at once oxidizes, and the 
reaction continues throughout the mixture in tlie crucible, the alumi- 
nmn readily taking up the oxygen from the metallic oxides i)resent 
and reducing them to pure metallic state. The heat generated by 
the reaction leaves the mass in a molten condition, the metals settling 
to the bottom and the alumina separating out as corundum. Metals 
such as chromium, titanium, tungsten, vanadium, molybdenum, etc., 
that it was formerly considered almost impossible to reduce to the 
metallic state, can now be obtained with comparative ease and at 
small expense. The corundum obtained varies in color and is 
dependent to a certain extent upon the metallic oxide reduced; thus 
in the reduction of chromic oxide the resultant corundum slag is 
ruby red; of nickel oxide, it is blue; of titanium oxide, it is brown; 
and of manganese oxide, it is yellow to greenish yellow. 
Within the last four years a process has been devised for manu- 
facturing corundum out of bauxite. This method consists essentially 
of subjecting the bauxite to very great heat in an electrical furnace. 
The idea that bauxite can be changed into corundum by intense heat 
is not new, for in 1801 Prof. T. Sterry Hunt, of the Geological Survey 
of Canada, in an article on " The origin of some magnesium and 
aluminous rocks," "' described the occurrence of bauxite at Baux, in 
the south of France, and stated that " by an intense heat this sub- 
stance is converted into crystalline corundum resembling emery in 
its physical character." 
Corundum is now being made commercially from bauxite by the 
I Norton Emery Wheel Company at their plant at Niagara Falls, N. Y. 
!The principle of their method is the heating of bauxite in a specially 
I constructed furnace about 4^ feet in diameter, using some of the 
i bauxite itself as a lining for the furnace. Before the bauxite is added 
I to the furnace it is thoroughly heated, in order to drive off not only 
I the moisture, but also the water of composition, so that the product 
I actually added to the furnace is a nearly anhydrous aluminum oxide. 
The heat generated is sufficient to fuse the bauxite, which recrystallizes 
I as corundum. The pig as it is taken from the furnace has an outer 
j layer of nearly unaltered bauxite. A portion of the corundum is 
iwell crystallized in hexagonal plates, which are very similar in form 
to those obtained in the reduction of the metallic oxides, as described 
; above. 
<» Am. Jour. ScL. 2d ser., vol. 32, 1861, p. 288. 
