﻿Vol. 66.~] GEOLOGICAL STRUCTURE OF SOTTIHERN" RELODGSIA. 373 



"to do more than say that 'blue ground' undoubtedly exists and 

 carries diamonds. The following remarks, therefore, apply to the 

 locality nearest the Bembezi River, now known as the Colossus 

 Mine. It appears to have an area of about 900 by 500 yards, and is 

 thus the biggest known deposit of its kind, considerably exceeding 

 in superficial area even the great Premier Mine in the Transvaal. 

 The walls of the pipe are granite, except on the north, where dolerite 

 and a green decomposed ' peridotite ' occur. The ' blue ground ' 

 presents the usual features, having a serpentinized matrix with 

 much calcite, and showing numerous rock-fragments and grains of 

 minerals such as garnet, ilmenite, augite (' chrome-diopside ') etc. 

 The rock-fragments are chiefly granite and dolerite, derived from 

 "the walls ; but there may also be noted inclusions of ' ultrabasic ' 

 rocks, among which are representatives of the ' eclogites ' that have 

 given rise to so much controversy. I have not yet been able to 

 make a detailed study of these inclusions, but append brief descrip- 

 tions of two leading types : — 



A. — la hand-specimens, shows a dull-green cleavable mineral forming a 

 kind of matrix, in which are embedded numerous small orange-coloured 

 garnets. Under the microscope the garnets appear as colourless isotropic 

 grains, with some tendency to crystal outline. Augite, colourless in section, 

 occurs in somewhat greater amount, making up nearly all the rest of the 

 rock, apart from a little yellowish, not very pleochroic mica, usually seen 

 along the edges of the garnets, but never forming a complete rim. 



B. — A dark, nearly black rock, showing occasional small pink garnets. 

 Under the microscope it is seen to be a coarsely granular aggregate of olivine, 

 enstatite, augite, garnet, and mica — in that order of abundance. The olivine 

 is very fresh, with occasional lines of separated nugnetite and very little 

 yellowish or bluish serpentine. The enstatite is colourless, and most of the 

 grains are somewhat elongated in the direction of the vertical axis, showing 

 well-marked cleavage-traces, parallel to which extinction takes place. The 

 augite is also quite colourless where unaltered, but most of it shows partial 

 jconversion into rather dirty-looking, but not distinctly coloured uralitic horn- 

 blende. Several garnets, from less than 1 millimetre to about 2 mm. across, 

 are seen in the slice. They are colourless and quite isotropic, and are 

 bordered wholly or in part by flakes of yellowish-brown mica. These latter 

 have no definite orientation towards the garnet, the cleavage being tangential, 

 radial, or at any angle. There is often a brownish band of alteration-products 

 between the mica and garnet. There are rare flakes of the same mica away 

 from the garnets. 



Personally, I do not believe that the occurrence of diamonds has 

 any connexion with the ' eclogites ' or any other inclusions found 

 in the various pipes. 1 The ordinary garnet of the ' blue ground' 

 itself is always a deep red pyrope, quite different from those of the 

 eclogites. Moreover, garnets are exceedingly rare at some localities, 

 notably the large and rich Premier Mine of the Transvaal ; and their 

 abundance is no guarantee that diamonds are present, as witness so 

 many k blank pipes ' all over South Africa. A notable amount of 

 chromium seems to be characteristic of the garnets of * blue ground.' 

 The following analysis of material obtained by me from the Colossus 



1 See Rep. S.A. Assoc. Adv. Sci. 1908, pp. 105-106. 



