52 2 Dr. Schafhaeutl on the Different Species of 
nevertheless a quantity of azote^ it must have been taken up 
during the process of puddling. 
Silicon is also a necessary ingredient in all sorts of cast 
iron ; no carhuration of iron would take place without its 
presence, and the iron appears only to have dissolved the 
carburet of silicon. Silicon shares this property of carbon- 
izing iron with manganese, and both are in some degree 
equivalent to each other, so that in certain cases the action of 
manganese is substituted for that of silicon. 
Such iron, in which manganese is substituted partially for 
silica, has that peculiar property of producing, in a certain 
species of refining fire, steel instead of malleable iron ; it is al- 
ways obtained from decomposed spathic iron ores; its fracture 
is silver white, and possessing a highly large laminous cry- 
stalline structure, and is the well-known spiegeleisen of the 
Germans. 
To give the reader an idea of the above-mentioned curious 
fact, I refer to the analysis of four sorts of white cast iron, 
made by the celebrated chemist Gay Lussac. The first speci- 
men is white, very similar to the other three. It produces ex- 
cellent malleable iron, but very inferior steel in the direct way. 
Iron from Champagne. 
De L’Iser. 
Siegen Coblentz 
in Germany. 
Carbon 
. 2-324 
2-636 
2690 
2-441 
Silicon 
. 0-840 
0-260 
0-230 
0-230 
Phosphorus .. 
. 0-703 
0-820 
0-162 
0-185 
Manganese 
, traces 
2-137 
2-390 
2-490 
Iron 
. 96-133 
94-687 
94-328 
94-654 
We see in the first specimen of the cast iron of Champagne, 
which produces excellent malleable iron, 0*840 of silicon with 
only a trace of manganese. In the second specimen, on 
the contrary, only 0*260 per cent, silicon are to be found, but, 
instead, manganese amounting to 2*137 per cent. 
If we consider the quantity of silicon in the first example 
as a standard, the three other specimens contain only 0*26, 
and therefore 0*58 silicon less ; now, 0*58 silicon wants 0*627 
parts of oxygen to become oxidized to silica, but the quantity 
of oxygen which is necessary to convert the 2*137 of manga- 
nese into protoxide is 0*615 ; in fact, almost the same as for 
the silicon of the weight 0*58 before-mentioned. 
This quantity of manganese gives to such iron the property 
of producing steel immediately in the refining fire, instead of 
malleable iron ; and the explanation of this enigmatical pro- 
perty is given by the following facts, of which I shall treat 
more largely in another place. 
