THE ORIGIN AND FORMATION OF THE DIAMOND. 173 



specific gravities of molten iron and diamond. Tlie specific 

 gravity of molten iron is over 7, that of diamond 3"5. If, then, 

 the diamonds in the South African Mines had been formed in 

 molten iron, we should expect that as soon as a diamond of, say, 

 two or three carats' weight were formed, that is, as soon as a 

 stone had attained a size sufficient to enable it to overcome 

 the viscosity of molten iron, it would rise to the surface of the 

 molten mass, and thus be to a considerable extent removed from 

 its sphere of action. This brings us to the question as to the 

 conditions under which such masses of molten iron were supposed 

 to be produced. Masses of molten iron deep down in the earth, 

 as supposed by this theory, would be pretty much in the same 

 condition as molten iron in a blast furnace, covered over by a 

 certain thickness of molten olivine rock : because, if the iron 

 were at a high temperature, the rock, containing a certain 

 l^roportion of iron silicates, which lower the melting point — would 

 necessarily also be in a molten condition. The specific gravity 

 of such a rock would be near that of the diamond, olivine being 

 3'3 to 3*5, and garnet from 3 to 4'5. If, then, according to this 

 theory, masses of molten iron formerly existed deep down at the 

 roots of the diamond mines, their temperature was such that 

 they were probably overlaid by olivine or other similar rock, 

 also in a molten condition. Under such circumstances it would, 

 I submit, be impossible for diamonds of such large size as are 

 frequently founcl in the mines to be formed, because as soon 

 as they had attained even a few carats in size their buoyancy 

 over that of the molten iron in which they were immersed would 

 cause them to float to the surface, and probably also to pass into 

 the molten rock above. Once out of the sphere of action they 

 would not only cease to grow, but they would in turn be them- 

 selves attacked by the molten rock, and destroyed in the way 

 indicated by Leipsic experiments. It has also been objected 

 to the molten iron theory that only minute diamonds could be 

 formed in that way, because, in the experiments made on those 

 lines, the diamonds only began to form at the moment of solid- 

 ification of the iron. I have not, however, made use of that 

 objection, because it seems to me if there is anything in the theory 

 that its advocates may claim that the diamonds were not 

 formed at the moment of solidification only, but that they were 

 formed by the excess of carbon crystallizing from the molten 

 iron as the latter cooled down from a very high temperature to 

 the solidifying point. There are other objections to the molten 

 iron theory, some of which have been noted by Mr. Gardner 

 Williams in his book on the Diamond Mines of South Africa ; 

 and I quite agree with him as to the difficulty, if not impossibility 

 of diamonds of any size being made by this method. 



Destructive criticism in this, as in other matters, is as we know, 

 generally easier than to provide a sufficient explanation of the 

 difficulties of the problem, so we may now very naturally ask 

 for some other theory in place of those rejected. If diamonds 

 were formed in the mines when the contents were in a mudlike 

 condition can w^e give an acceptable explanation of the modus 



