iron . 
447 
oxygen, thus becoming more difficult to be reduced than ever. 
This is indicated by its swelling in bulk and losing its magnetic 
virtue. Upon this supposed absorption of oxygen, unsupported 
as it is by any experiment that I know of, Mr. Mushet’s whole 
theory of torrefaction is erected. 
Now with respect to the fact that ironstone over roasted, loses 
its magnetic property, it is true only as it regards the presence of 
sulphur. I have repeatedly torrefied the fine argillaceous iron- 
stone of Bassenheim iron-works, from a low red to a fusing heat, 
and found its magnetism in the last stages increase instead of dimin- 
ishing. The result was very different however, when the ore 
contained sulphur. It then fused much sooner, became porous, 
and spongy, and lost its magnetism altogether. If it were oxygen 
that produced the effect in the one, it ought also to have produced 
it in the other. The obvious inference must be, that Mr. Mushet 
erected his theory upon the examination of sulphureous ironstone, 
frithout observing the difference ; since indeed he not only ne- 
glects taking into consideration the effects of sulphur generally, 
but even ridicules the folly of those manufacturers who consider 
its agency of such importance. 
But there is another powerful reason against this supposed 
increase of oxygen, derived from the very principles of the ore 
itself. For ironstone can in most cases be considered in no other 
light than a peroxyde of iron ; already subjected as it most probably 
has been, to oxydation, during centuries, —it must long since have 
reached its point of saturation, and therefore might part with, but 
assuredly cannot acquire oxygen during the operation of roasting. 
To ascertain this matter however beyond all dispute I took 
some of this sulphureous ironstone which had been torrefied till it 
fused into a light cellular mass of a dark livid blue colour, totally 
destitute of magnetism. This I pulverized, and having again sa- 
tisfied myself that not an atom of it was attracted by the magnet, 
I put it into a glass with some sulphuric acid. An immediate action 
was perceived and a smell of hydrogen emitted. I diluted the acid 
and let it stand for a couple of hours, when on testing it by prus- 
siat of lime, a greenish blue precipitate instantly apppeared. It is 
hardly necessary to mention to the chemical reader, that the prus- 
siat of lime here employed, is preferable to the prussiat of potass 
as a test, unless this last be very carefully prepared, because the 
prussiat of potass as commonly used is apt to be decomposed 
where there is an excess of acid, but the prussiat of lime (unless 
