274 GEOLOGICAL SURVEY 
economical point of view, it is said that, for this latter process, dry wood, branches of trees, &c. 
may be used. 
We shall now be better able to see the force of Mr. Henderson’s experiments. He exposed frag- 
ments of the ore, in contact with pure charcoal, in a crucible, excluding atmospheric air, for the 
space of thirty hours, to a red heat. He next filed and polished the pieces, when two different 
colors were observed on the polished surfaces—the one dark and dull, the other bright and metalic. 
Moreover, the dark portions were found to be brittle, but were acted on by the magnet, while the 
portions having the bright lustre were malleable, being truly metallic iron. 
To account for these changes, we must again recur to what has already been said, in relation to 
the fact that magnetic iron is a mixture of the protoxide and peroxide. In the experiment the for- 
mer parted with its oxygen, and was consequently reduced to the state of malleable iron; whilst 
the portion of the ore consisting of the peroxide also gave up a part of its oxygen, and was con- 
verted into the protoxide, which appeared in the specimens as the dark magnetic portions. 
It appears, still further, that the different varieties of ores were made up of very different propor- 
tions of these oxides; the Mack ores being converted into metallic iron, with, however, small black 
portions disseminated throughout: the metallic portions representing the amount of protoxide, and 
the black, disseminated specks or crystals, the proportion of the peroxide in the specimen. On 
the other hand, the fine-grained compact ores showed a large proportion of the peroxide, for the 
dark portions of the polished surfaces greatly exceeded the parts converted into malleable iron, 
which represents the protoxide. 
From these experiments of Mr. Henderson it also appears that if the process be continued until the 
whole be reduced to the metallic state, after the peroxide is reduced to iron, being in contact with 
charcoal, while the peroxide is parting with its last atom of oxygen, the former oxide combines 
with carbon, and is converted into steel, so that the mass will be a mixture of steel and iron. 
From this brief account of these experiments, it seems to me that the cause of much of the diffi- 
culty experienced in working the magnetic ores will be obvious. It points out, too, the reason why 
the compact varieties of these ores, although absolutely richer than the others, are not equally valued 
at the works. 
It seems also that, to a certain extent, these ores contain, in their composition, a remedy for these 
difficulties, for it appears that it is the protoxide alone that is affected by the magnet; and hence we 
have, from this quality, the means of separating the two oxides by the magnetic machine. 
It has been proposed to roast these ores, and by this means to convert a portion of the peroxide 
into the protoxide, by causing it to part with a portion of its oxygen. The ores are also found to 
be more magnetic after roasting. After being pulverized, the ore is passed through the magnetic 
machine, which consists of a cylinder about two and a half feet in diameter, and five feet in length, 
studded with bar magnets on the insides. As the ore passes over these, the protoxide is collected 
and retained, while the peroxide passes off with the impurities. The ore is brushed from the mag- 
nets into a trough placed to receive it. Such a machine as this is capable of separating five or six 
tons a day. 
The impurities with which the ore is mixed, are best removed by washing, after the ore is pul- 
verised, by passing a stream of water over it. The stream must not be too rapid, otherwise much 
