20 
Transactions of the Boyal Society of South Africa. 
veinlets of a soft rather asbestiform mineral, which is probably a form of 
bastite. 
Pebbles of a remarkable porphyritic rock occur rather abundantly 
in two places in the Umtebekwe river, but the rock was not found 
in situ. This consists of a fine-grained pale grey-green pyroxene-felspar 
ground-mass, in which are embedded large perfectly idiomorphic white 
felspar crystals up to several inches in size. The rock probably belongs 
to the Great Dyke. 
V. — Conclusions. 
The evidence obtained during the examination of the Selukwe section 
with regard to the habit of the intrusion is meagre, in large part negative 
and to some extent contradictory, and it must be borne in mind that the 
habit may be disguised in several respects and probably cannot be made 
out as a whole from the examination of a portion. Other districts may 
afford definite evidence from which conclusions on the point may be 
drawn w^ith greater certainty. It is thought, however, that at Selukwe the 
balance of the evidence is in favour of the intrusion having the dyke 
habit, but that it is likely to be a dyke with rather marked hade. It 
is therefore proposed to retain the name "Great Dyke" for the present. 
As regards the mineralogical composition, texture, and distribution 
and proportions of minerals in the several rock types, remarkably stable 
conditions appear to have been reached before consolidation of the 
component parts into which the magma split. Thus throughout the 
intrusion the rocks are evenly medium to somewhat coarse-grained, 
and the constituent minerals are fairly evenly distributed through the 
rock, even in cases where there is but 10 to 15 per cent, of one of 
the essential constituents. The rock types further are sharply differen- 
tiated ; it is exceedingly rare for one type to grade into another. 
The intrusion is built up of a consanguineous series, but w^ith the 
evidence at present available the matter does not permit full discussion 
of the origin and differentiation ; nevertheless it may be useful to consider 
the facts ascertained in detail and attempt to draw conclusions from them. 
The differentiation does not appear to have been one simple gradual 
operation. It is assumed that the parent magma split into two partial 
magmas, one with olivine and little felspar, the other without olivine 
and with much felspar, but both with enstatite ; the less basic magma 
came to occupy a central position. Each of these partial magmas it is 
assumed again separated into portions respectively rich in, and com- 
paratively poor in, enstatite. The extremes are a non-felspathic peridote 
enstatite rock and a non-peridotic felspar pyroxene rock. 
The following diagram shows hypothetically the origin and relationships 
of the types. 
