83 
1912-13.] Magnetic Induction in Ferric Oxide. 
The above experiments were carried out in the Natural Philosophy 
Institute of Glasgow University, and the authors desire to thank Professor 
Gray for his aid and encouragement during the progress of the work. 
Chemical Transformation of Ferric Oxide when continuously 
HEATED IN A REDUCING GAS. 
Moissan * appears to have been the first to make a complete investiga- 
tion of this question. He states that when ferric oxide is heated in pure 
dry hydrogen or carbon monoxide it is reduced, the changes following the 
order : ferric or sesqui-oxide, Fe 2 0 3 — magnetic or ferrosoferric oxide, 
Fe 3 0 4 — ferrous or protoxide, FeO — metallic iron. The change from Fe 2 0 3 
to Fe 3 0 4 takes place between 350° C. and 440° C., and the magnetic oxide 
obtained is an allotropic form of that produced at high temperatures, e.g. 
that found in meteoric dust. The magnetic oxide when further heated at 
500° is reduced to ferrous oxide. This reduction is more complete when 
done in carbon monoxide than when done in hydrogen, and the protoxide 
obtained is free from magnetic oxide and from metallic iron — a result 
difficult to attain when hydrogen is employed. When the protoxide is 
heated at 600° to 700° in hydrogen, it is reduced to metallic iron. 
If the temperature is kept constant, then, under the same conditions, 
the protoxide is formed at 440° in about 6 hours, and metallic iron after 
12 hours. 
Hilpert,f who repeated some of Moissan’s work, found that the 
temperature of reduction from Fe 2 0 3 to Fe 3 0 4 depended in great measure 
on the temperature to which the ferric oxide had been previously heated, 
and that reduction began at temperatures which varied between 280° and 
350°. The reduction does not proceed at the same speed throughout the 
material, and preparations may be obtained in which all stages of reduction 
are present, from unchanged red oxide to metallic iron. 
Other observers J have found the reduction of ferric oxide to metallic 
iron to take place at temperatures as low as 370°. 
In addition to these observed differences in chemical behaviour, all three 
iron oxides have been described by chemists as existing in two allotropic 
* Ann. Chim. et Phys., v. 21, pp. 199-255, 1880. 
t Ber. d. chem. Ges., xlii. 4, pp. 4575-4581, 1909. 
I Siewert, Jahresber. Chemie, pp. 265-266, 1864 ; Glaser, Zs. anorg. Ghemie, xxxvi. 
p. 21, 1903. 
