132 
Transactions of the Boijal Society of South Africa. 
and the terraced stones of Jagersfontein, groups and clusters of diamonds 
are by no means uncommon. These groups and clusters consist sometimes 
of diamonds of apparently perfect symmetry, sometimes of irregular lumps; 
and they may be all of one colour or there may be two or three different 
colours and textures. A brief description of one or two will suffice : 
(1) A close cluster of separate diamonds, of which upwards of fifty can 
be counted on the outside. Originally it must have contained many more 
than it does now, for it is in so fragile a state that individual components 
are continually breaking away. Indeed, it is a wonder that it ever reached 
the sorting office as a cluster at all. The components are variously coloured 
from white to grey. Those that break off have fractured surfaces as well 
as surfaces of arrested growth. 
(2) A crystal diamond surrounded by a cluster of lumps of bort. 
(3) A branching group of five tiny dark grey diamonds, possibly coated 
stones. They seem to be rhombic dodecahedra, and are what is left of a 
larger group. 
(4) A beautiful group of three tiny white diamonds, two of which are 
well-shaped octahedra and one rounded. 
(5) An irregular group of ordinary grey bort of twenty or more com- 
ponents of irregular shape and of various shades of grey. 
(6) A Bultfontein cluster of yellowish diamonds of different habits. 
The principal member is a made of 8*7 x 5"1 mm., one of wdiose seamed 
sides has developed into an octahedral face. Another side and one corner 
are distorted where a satellite diamond has broken away; while a well- 
shaped made, of the same thickness but smaller and not lying quite in the 
same plane, has grown into the third side. Close to the second made is a 
distorted octahedron, one of whose corners shows where there was once 
another diamond. Wedged between the second made and the distorted 
octahedron are two dodecahedra — there seem to have been three originally. 
This list could be extended indefinitely ; and considering that the 
chances are against a group or cluster surviving intact the ordeal of mining 
and winning it seems a fair inference that groups and clusters must be very 
common in the original matrix, and that it is immaterial to a diamond 
whether it forms individually or socially. In fact, saving lack of material, 
there is no reason why a group of diamonds should not extend from one side 
of the mine to the other. 
It might be argued against this view that possibly the components of a 
group have started from closely situated independent centres of crystallisation 
and coalesced by natural growth. Without going so far as to assert that 
such a thing had not happened and could not happen, it may yet be said 
that chances and appearances are against any such phenomenon. Moreover 
the case of the Bultfontein spinel twin made described and figured above^ 
also the case following it of the Koffyfontein diamond, show in a way how 
