516 Dr, Schafhaeutl on the Different Species of 
assume that the metallic iron is in combination with carbon, 
azote and hydrogen, or perhaps in a state forming a sort of 
cyanet of iron. 
The relation of the quantity of constituents of the brown 
residuum of the iron is so stable and so accurately in atomic 
proportions, that we may safely conclude the quantity of carbon 
from the amount of oxygen absorbed. 
If we denote the increase of weight by letter (a), the oxygen 
of silicon as ( 5 ), the oxygen of iron by (/*), and the carbon 
by (a**), we obtain always the following simple formula, par- 
ticularly when we consider that the iron is oxidized according 
. 3 ... 
to the above-mentioned formula F F : 
= 5 -f f —a. 
In some French irons I found the relations of carburets of 
iron to the siliciuret approach 
S C^. 
F4 QA 
If we further consider, that the white silica only remains 
when the iron residuum has been exposed to a number of ig- 
nitions, and that after only a few ignitions it remains always 
dark black, soluble neither in alkalis nor acids ; we may find 
therein likewise a proof, that silicon is combined chemically with 
carbon. Further, the remainder of the residuum five times 
ignited had gained 0*028 grains, after the black-coloured 
scales had disappeared ; this increase in weight can only be 
ascribed to the oxidation of those black scales, which must have 
been a metal whose oxide was white. Further, if we consider 
that the remainder of the residuum only once ignited, after 
having changed its deep-black colour to white, and after emit- 
ting a great quantity of smoke, had only lost 0*05 11 grains; 
v/e may take all these circumstances as a certain proof that 
the silica obtained from these residuums must have been con- 
tained in the metallic state, and in combination with carbon, 
which is likewise to be suspected from its emitting bright red 
sparks when thrown into the flame of a candle. 
Having now considered partially by induction the relative 
combination of the ingredients of the residuums of white 
irons, we proceed to ascertain by experiment the quantity 
of the volatile bodies in the residuums. 
60 grains of the same Maesteg white iron, freed from all fine 
powder, and treated in a retort with 4 oz. of hydrochloric acid 
of sp. gr. 1*108, generated no spongy foam whatever; the li- 
quid in the retort always remained transparent, and the action 
of the acid had ceased the next day. 
13*711 grains of this residuum, corresponding with 35 grains 
