1899] THE SOUTH AFRICAN DIAMOND 181 
eclogite, as a zircon is of a granite or syenite. It may prove, how- 
ever, not to be restricted to this one species of rock. I see no reason 
why it should not also occur in the enstatite-eclogite already described ; 
while the fact that at Kimberley, if not at Newlands, olivine is 
abundant in the diamantiferous blue ground stegests the possibility 
that the diamond may also be a constituent of a peridotite. In fact, 
though I was unable to accept my late friend Professor Carvill Lewis’s 
view that the Kimberlite was an altered peridotite, I fully expected 
that sooner or later it would be traced back to some very basic rock, 
probably to a peridotite. ‘lhe diamond hitherto has only been proved to 
occur in meteoric iron! (Cafion Diablo), and it was made artificially 
by Professor Moissan by the intervention of that metal. Indeed, on @ 
priori grounds I should have expected to find it in a rock less acid than 
an eclogite. I venture, accordingly, to suggest that the crystallisation 
of the carbon may possibly have occurred in some very basic magma 
which was afterwards invaded by one more acid, the eclogite being the 
result of the mixture. This, however, is a speculation; the fact, I 
think, cannot be disputed that the diamond has been traced back to an 
igneous rock (eclogite) and was not formed in the “blue” (Kimberlite). 
The boulders described above appear to me truly water-worn ; so also 
are not a few of the smaller fragments. I suspected this some time 
ago when examining a parcel of “ washings” from the De Beers Mines 
(where also boulders have occurred), but those sent to me from 
Newlands have placed it beyond doubt; half a small pebble of 
eclogite is present, while many of the minerals are so well rounded 
that the darker kinds could only be determined by fracture. But if 
this be so, if many of the constituents are water-worn, how can the 
so-called Kimberlite be an altered porphyritic peridotite? We are 
compelled to regard it as a clastic rock, formed by explosions, which 
-have mingled the shattered constituents of the coarsely crystalline floor 
with materials derived from the overlying sediments. The comparative 
abundance of diamonds in the blue ground suggests that they are fairly 
common in some members at least of the holocrystalline series. Hence 
it may be possible, by carefully observing the larger minerals found 
with diamonds, to infer which of them are really its associates. At 
present, garnet, chrome-diopside, and perhaps iron oxides, can alone be 
named, but I fully anticipate other pyroxenes and olivine to be added. 
Hence, as the blue ground is not an altered peridotite, the 
name Kimberlite must be removed from the list of that group, and 
must disappear from science, unless it be retained for this peculiar 
breccia in which the diamond very commonly is an accidental con- 
stituent. The mode of occurrence, structure, and contents of this 
breccia suggest that it is the result of some kind of volcanic action, but 
the general absence of scoria makes it probable that the explosions were 
due to accumulated steam, and were thus of an exceptional character. 
1 The Novo Urei meteorite, however, is said to contain some ferro-magnesian minerals. 
