THE HARDENING AND TEMPERING OF STEEL. 
251 
has generally been called ‘ free 9 carbon, and when in solution 
‘ combined 9 carbon. As to the proportion in which these two 
forms occur in various qualities of iron and steel, Akerman, 
Barba, and others, have advanced various opinions ; but the 
view that there exists in. steel or cast-iron any definite chemical 
combination, or carburet of iron, appears to be abandoned on all 
sides. It seems generally allowed that no other elements than 
carbon and iron should be present in really pure steel ; and 
that other substances, such as manganese, which are generally 
considered to improve steel, only do so by neutralizing the 
action of other and more hurtful impurities. 
II. Condition of the Carbon of Iron in Steel. — Jullien’s view on 
this part of the subject is, that molten cast-iron is a solution of 
liquid carbon in liquid iron : that when the metal is cooled 
slowly, part of the carbon separates out as pure graphite, while 
the remainder continues in solution, thus forming ‘ grey/ or 
soft cast-iron ; but that, when the metal is cooled quickly, this 
separation does not occur, and that the result is a solution of 
crystallized carbon in amorphous iron, forming ‘white/ or 
hard cast-iron. Exactly the same process occurs with steel, 
and makes the difference between hard and mild steel. Caron 
and Akerman generally confirm this view. They point out 
that hardened steel dissolves in hydrochloric acid without leav- 
ing any residue ; but that the same steel, if first annealed, by 
being kept a long time at a red heat and allowed to cool slowly, 
leaves a residue of carbon insoluble in the acid. This seems to 
show that there is a more intimate mixture between the two 
elements in the former case, and that the carbon either dis- 
solves with the iron, or, more probably, escapes as carburetted 
hydrogen. 
III. Hardening of Steel. — On this, the central point of the 
whole inquiry, Jullien advances a very bold and original theory. 
He holds that carbon, in contact with red-hot iron, becomes liquid 
and is absorbed like water in a sponge ; that, if cooled slowly, 
the carbon becomes amorphous, and the steel is soft ; but if 
cooled quickly, the carbon crystallizes, taking the properties of 
diamond, and the steel becomes, in fact, diamond set in iron. 
The hardness of the steel is thus simply due to the hardness of 
the crystallized carbon. In support of this, he remarks that all 
hard bodies take different molecular structures, according as 
they are cooled rapidly or slowly ; e.g ., gold is fibrous when 
cast in a metal mould, crystalline when cast in a sand mould ; 
glass is crystalline when cooled rapidly, but amorphous when 
annealed. He further observes that diamond, heated for a long 
time in a closed vessel, becomes graphite, and hence concludes 
that liquid carbon, rapidly cooled, would become diamond. 
This theory, striking as it is, is beset with difficulties. As 
