TRANSACTIONS OF THE SECTIONS. 67 
water or oxygen when in the pan, must have formed an extremely vo- 
latile combination, evaporating as rapidly as it was formed. By a 
repetition of this treatment with the acid, graphite in the third stage, 
as before described, was obtained ; the metallic lustre was entirely lost, 
as also the laminated texture; it now appearing as a porous mass, re- 
sembling coke, and no longer capable of reduction to its original 
foliated state by pressure, having undergone a decided chemical change. 
The acid solution deposited at once a copious precipitate of silica and 
alumina (slightly tinged by oxide of iron), upon the addition of ammo- 
nia to neutralization. A similar precipitate was obtained from the 
acid of the first experiments, but less in quantity, and requiring a 
longer time for its operation. Thus it would appear that the abstrac- 
tion of silicon and the change of physical properties of graphite, are 
corresponding and mutually connected phenomena. In further proof 
of the necessary connexion of silicon as a chemical combination, essen- 
tial to the existence of the scaly metallic lustre of graphite, it will be 
found that by repetition of the same experiments, the globules ultimately 
disappear, and the remaining solution in acid neutralized by ammonia, 
deposits only flaky silica, with traces of oxide of iron. On observing 
attentively the specimen of graphite, as found in its natural state, and 
comparing it with those treated with acids and alkalies, also exhibited, 
it appeared that the scales, before being operated on, had a dirty gray- 
ish appearance, described as owing to their being covered with spots, 
consisting of microscopic six-sided flattened prisms of silicate of iron ; 
the matrix of this graphite formation, in the blast-furnace cinder, es- 
sentially composed of bisilicate of lime and alumina, deriving a yellow- 
ish tint from a slight admixture of sulphuret of calcium, with a trace 
of sulphuret of potassium. Scales of very different density may be 
separated, the thinnest unaffected by the magnet, the thicker ones 
decidedly so; those in the middle of the mass, thicker and stiffer, not 
easily broken, and showing a shining black fracture, like that of an- 
thracite, form a variety of graphite, in which silicon and iron are 
greatly predominant, developing when treated with hydrochloric acid a 
fetid hydrogen characteristic of cast iron, and separating at the same 
time yellow flocks of silica and alumina. Dr.Schafhaeutl then proceeded 
to point out an analogy between the formation of gray iron in the blast 
furnace, and that of graphite; namely, that the same chemical condi- 
tions occur during the change of white iron into gray ; this takes place 
after having descended through the furnace, and reached the stratum 
of slag covering the melted metal; this slag being an earthy bisilicate 
(in coke furnaces approaching to a trisilicate), and containing a small 
quantity of protoxide of iron. As silicon is found in graphite only in 
very small quantity, it has been considered an accidental impurity, just 
as the small quantity of hydrogen retained by charcoal, sulphur, &e. 
has been considered an impurity; but as these foreign matters can by 
no chemical means be separated, without destroying the state in which 
graphite, charcoal and sulphur exist, it must be inferred that such 
admixture is essential to their existence in that state in which they 
ordinarily appear. Quitting now the individual consideration of gra- 
F2 
