cvili The Transactions of the South African Philosophical Society 
Very little recent work has been done in the better-known parts of 
the country. Mr. Sawyer has investigated the geology of the Prince 
Albert gold-fields, Mr. Duff that of the Knysna; and my own travels 
and researches have enabled me to settle the question of the origin of 
the nitrates, which occur in several mountain ranges. 
It is to be regretted that the splendid opportunity offered by the 
numerous water-boring operations has been partly lost by not preserving 
all the cores, especially from localities where different formations had 
to be cut through. 
The curious rock, which everyone who refers to South African geology 
must mention, the Dwyka conglomerate, has received almost as much 
attention as the auriferous districts, it being now generally recognised 
as an undoubtedly clastic formation, and most observers who have 
studied it recently admit that it appears to have originated in a glacial 
period. If the conglomerate which fringes the southern edge of the 
Karoo, and the similar rock found on the Vaal and Orange Rivers, are 
identical, as Dunn assumes—and there is some probability of that being 
the case, as the latter contains the same kind of scratched stones, some 
of which I have gathered myself near Prieska, breaking them out of the 
rock—then we would possess a vast deposit formed during a glacial 
period and spreading almost over the whole of South Africa. 
One branch of geological study, which is more and more recognised 
as being of the greatest importance, owes its progress, as far as South 
Africa is concerned, entirely to one man, viz., to Professor Cohen, at 
present at Greifswald. 
He and his pupils have ascertained the mineralogical composition of 
a considerable number of our rocks, ¢g., granite from Table Mountain, 
clayslate and diabase from Sea Point, and a number of varieties of 
porphyry, gabbro, diabase, and granite from various localities in the 
Transvaal. The importance of the work done, and the urgent neces- 
sity for further research in this direction, is best shown by the fact that 
many of the rocks formerly called dolerite were recognised as diabase ; 
that the origin of many of the boulders occurring on the banks of 
the Vaal River near its junction with the Orange River was traced 
to the Makwassi Mountains; that the material which formed the 
Table Mountain sandstone was derived from granitic or gneissose 
rocks. 
It is the study of the petrography of our non-fossiliferous rocks which 
will be the principal factor in revealing their nature and origin and their 
true position to each other. 
What an interesting work would it be, ¢g., to carry out a complete 
petrographical investigation of the rocks and matrix of the Dwyka con- 
glomerate ! for that would indicate to us the regions whence all these 
