558 ME. D. DRAPER ON THE GEOLOGY [NOV. 1 894, 



defined ; but to distinguish the Ecca from the Beaufort Beds and 

 from the upper series is by no means an easy task. 



They all graduate one into the other, so that the line of demarcation 

 in many cases cannot be determined with any degree of certainty. 

 Fossil evidence is very scarce, except leaf-impressions, principally of 

 Glossopteris. These are found in the Molteno, Beaufort, and Ecca 

 Beds. No fossils have been found in the Dwyka Conglomerate in 

 any part of South Africa up to the present date. 



The ' Bed Beds ' yield an abundant supply of fossil reptilian 

 remains, and the Cave-sandstones yield specimens of fish. A. portion 

 of a fish was discovered by the writer in the Molteno Beds of Natal, 

 and is, I believe, the first specimen discovered in this series ; un- 

 fortunately, it is too fragmental for classification. Mr. A. Smith- 

 Woodward has described it in the Ann. & Mag. Nat. Hist. ser. 6, 

 vol. xii. 1893, p. 397, pi. xvii. fig. 4. 



Evidences of recent glacial action are entirely wanting ; but 

 Dr. Sutherland has noted ice-scratchings, etc., on the surface of 

 the rocks underlying- the Dwyka Conglomerate at Durban, and 

 possibly ice may have played some part in the creation of the 

 Dwyka Conglomerate. Ice is looked upon as the principal agent 

 in the removal of so vast a quantity of the sedimentary strata as 

 that now missing in South Africa by those who are in favour of a 

 ' glacial period ' having extended over the earth at some time of 

 its existence ; but the total absence of moraines, erratics, and other 

 evidences of glacial action must be accepted (as against vague 

 theories) in support of the arguments which tell against the 

 glaciation of the southern portion of the continent since the close 

 of the Jurassic period. 



I cannot imagine glaciers descending from the peaks of volcanic 

 rock without bearing boulders of the characteristic rock of that 

 series, and these would be deposited in the lower-lying valloys. The 

 absence of such boulders is strong evidence against any such glacial 

 theory. On the other hand, the softness of the rocks of the 

 horizontal strata, and the great amount of weathering to which they 

 have been subjected, in a climate with so variable a temperature as 

 the south-eastern portion of the continent possesses, would account 

 for the non-existence of any ice-scratches or markings. 



The numerous ' pans ' that are found all over South Africa have 

 been advanced as evidence of glacial action ; but here, again, 

 other evidence is entirely wanting. 'Pans' exist in South Africa 

 at all elevations, and in rocks of all geological periods — on plains, 

 in valleys, on the High-veld plateau, and on mountain-tops ; in fact, 

 almost every flat-topped hill in South Africa has a ' pan ' or small 

 lake on the top of it. They appear to be the result of the disinte- 

 gration and dissolution, by chemical means, of the rock-forming 

 constituents of the strata. 



Various products from the rocks in the neighbourhood are 

 generally found on the lower (or outlet) side of these pans — limonite, 

 kaolin, and clay-beds, derived from the destruction of the igneous- 



