xxxvi_ Historical Sketch of the Physical Sciences, 1818. 
7. Softening and tempering Steel—Mr. Gill informs us, in the 
Annals of Philosophy, xi. 58, that if steel be heated below the 
hardening point, and then plunged into cold water, it will be 
softened thereby, and in a much superior manner to the common 
process. He says, likewise, that if steel be heated to the requi- 
site degree, and plunged into a bath composed of a mixture of 
lead and tin, similar to plumber’s solder, heated to the requisite 
temperature, it will be at once tempered and hardened. It is 
for artists to determine whether these methods will answer, and 
whether they be preferable to the common ones. 
8. Manganese.—It is now about 44 years since this metal was 
reduced by Assessor Gahn, of Fahlun; yet I am not aware of 
any additional facts respecting it since that period, except those 
contained in Dr. John’s elaborate paper on this metat published 
in 1807. M. Fischer, of Schaffhausen, a manufacturer of cast 
steel, having discovered the means of producing a very intense 
heat in his furnaces, has been enabled in consequence to reduce 
this very refractory metal to the metallic state. The following 
are the characters of metallic manganese, as described by the 
editors of the Bibliotheque Universelle, to whom M. Fischer sent 
a specimen of the reduced metal. 
its colour is whitish ; it is harder than tempered steel ; it cuts 
glass nearly as well as the diamond; it scratches rock crystal. 
It acquires a very good polish, which is probably not durable, in 
consequence of its great affinity for oxygen. When kept under 
water for 24 hours, it becomes covered with a coat of brown oxide. 
It sensibly attracts the magnetic needle ; but was not, perhaps, 
quite free from iron. Its specific gravity is 7-467. This is con- 
siderably under the estimate of Dr. John, who found it 8-013. 
This I consider as an additional proof of the impurity of the 
manganese of M. Fischer.—(Biblioth. Universelle, vi. 232.) 
In the fifth edition of my System of Chemistry (vol. i. p. 403), 
I have endeavoured to show that manganese forms only two 
oxides, the green, and the black. The composition of these 
oxides, I consider to be as follows : 
ProtOwi Gena isisnia«s 100 manganese + 28°75 oxygen 
Peroxide s-0,6)a4¢ 100 manganese + 57:50 oxygen 
On this supposition an atom of manganese weighs 3°5. Ber- 
zelius had long before made a set of experiments on these oxides, 
and had determined their composition to be as follows: 
Protoxidg wh -.3s,«is .++e- 100 metal + 28°107 
Peromger sienna tions 100 metal + 56:214 
M. Arvedson has lately repeated the experiments of Berzelius, 
and obtained the same results. This certainly gives considerably 
additional weight to the determination of Berzelius, whose well- 
known precision entitles all his experiments to the greatest 
attention. My numbers were pitched upon from theoretical 
