THE ORIGIN AND FORMATION OF THE DIAMOND. 177 



molecules of the various forms of carbon must be very complex, 

 because if not, carbon with its low atomic weight ought to be 

 a gas. or at least a liquid, if we allow the possibility of its 

 tetrad valency modifying its state. There is the strongest reason 

 to believe that the elementary forms of carbon contain at least 

 six atoms to the molecule, and graphite and diamond probably 

 a multiple, or multiples of that number. Thus amorphous 

 carbon can be transformed into mellitic acid. Graphite also 

 can be partly transformed into mellitic acid by treating the residue 

 obtained from graphitic oxide with oxidising agents, though only 

 apparently after its molecule has undergone considerable degrad- 

 ation. The chemical deportment of diamond shows, as Briihl 

 points out, that it cannot have any double bonds in its molecule, 

 whereas graphite almost certainly has such, as is evinced by the 

 formation of graphitic o.xide, (S'c. So far, then, as theory may 

 help us in the formation of diamonds we may take those deductions 

 into account. The dil^culty with carbon is not to get the atoms 

 linked together but to prevent them doing so, or, rather to 

 prevent them doing so in the wrong way. My experiments, and 

 I have made many, many hundreds, have shown me that once 

 the carbon atoms are linked up into one of the modifications 

 of amorphous carbon, or into the form of ordinary graphite (please 

 note that I say ordinary graphite) it is most difficult, if not im- 

 possible, to transform them into another form, excepting of course 

 at a very high tem.perature, when no doubt the molecular structure 

 has a tendency to be broken up. Hence, in our experiments, 

 not only must we aim at forming a special kind of molecular 

 structure, but it is very important that the carbon atoms, as 

 they are liberated, shall be free to associate themselves in the 

 way we wish. And it is a curious fact that, in those experi- 

 ments which have succeeded in forming minute diamonds, the 

 carbon atoms, to begin with, have, so to speak, been free from 

 other entanglements. Even in molten cast iron the carbon 

 atom may exist unconnected with other carbon atoms, as metal- 

 lurgists tell us it exists as FcsC in the solid iron. The formation 

 of minute diamonds by the electrolysis of chloride of carbon, 

 and by the action of sulphur on carbide of iron in closed vessels. 

 makes use of carbon compounds in their sim]:)lest form. I have 

 thought, also, that the stage at which the specific heats of the 

 three modifications of carbon approach each other, viz., at a 

 temperature near 600° C, may have an important bearing on 

 this matter, as undoubtedly some intramolecular change then 

 takes place. As we know graphite is obtained at a certain stage 

 in the manufacture of caustic soda, and is supposed to result 

 from the decomposition of some ferro-cyanide of sodium present. 

 It is also said to be present in the black residue resulting from 

 the decomposition of pure hydrocyanic acid. Such compounds 

 are endothermic, and it may be supposed that at the moment 

 of decomposition the atoms are for an instant raised to a high 

 temperature, though the temperature of the mass does not rise 

 much. At one time I experimented much with acetylene gas, 

 dissolved in water rmder a high pressure, as such a gas might 



