of Edinburgh, Session 1880-81. 
21 
calcining sugar) is then reduced to an impalpable powder in an 
agate mortar, and laid in alternate layers with the metal (in small 
lumps) in the carbon crucible as already described, great pains being 
taken to secure that the 
metal is well surrounded 
with carbon. The carbon 
lid is then tightly fitted, 
and any little crevice filled 
up with pure charcoal. 
The graphite crucible lid 
luted on with the mixture 
of carbon and gum and 
the whole again carefully 
dried. This graphite cru- 
cible is then placed in an ordinary steel melting crucible (to facili- 
tate its being handled) and surrounded with charcoal and coke-dust. 
The whole then placed in a Siemens’ regenerative gas furnace and 
kept at the temperature of melting steel for from nine to ten hours. 
It is then taken out and the crucible buried in hot sand, to allow 
of very gradual cooling and thus give every chance for the formation 
of crystals, in this way it was found possible to extend the cooling 
over from fourteen to eighteen hours. 
During the heating the melted metal becomes thoroughly saturated 
with the carbon, which again crystallises out on cooling. On open- 
ing the crucible the metal is found in a single lump towards the 
bottom, and still surrounded by the undissolved carbon, from which 
it comes away quite easily and cleanly, and only requires to be 
washed with water and a small brush to remove the adhering 
carbon, when it is ready for extracting the crystallised carbon from 
its interior. On examining the metal at this stage of the operation, 
marked lines of crystallisation can be distinctly seen crossing the 
silver in two directions at right angles to one another. 
The metal is now dissolved in nitric acid, when we obtain the 
dissolved carbon from its interior in the form of a greyish black 
powder possessing a beautiful graphitic lustre. On examining this 
powder under the microscope we find it to consist of three different 
kinds of substances — first of graphite, which forms the larger pro- 
portion ; secondly of a number of small crystalline bodies of 
