30 Transactions of the South African Philosophical Society. 



Neck, to ones which must have had a molten magma like the 

 Nightingale agglomerate. 



In the Drakensberg I found a great many of this kind of volcano, 

 and also many filled in with a peculiar white chalky-looking sand- 

 stone. Just under the lavas which form the crest of the Drakensberg 

 there is a thick bed of sandstone of precisely similar nature, called 

 by Dunn the Cave Sandstone. One would at first be tempted to 

 think that the explosion that caused the vent, burst through this 

 sandstone, and the shattered remains then fell back and filled the 

 pipe. Against this, however, we must remember that the Cave 

 Sandstone is a comparatively thin band among a vast thickness of 

 beds made up of other materials, and it is not probable that the 

 rending of the outburst would affect this particular bed alone ; if the 

 Cave Sandstone was formed before the production of the material in 

 the pipe, then the latter ought to be made up of portions of all the 

 rocks which the vent traverses, but we find that this is not the case, 

 and that the material in the pipe is identical with that of the Cave 

 Sandstone. We must, therefore, suppose that the Cave Sandstone 

 was produced as a tuff which was blown out of the vents, a portion 

 of which still, in some cases, remains in the throat. 



I was greatly puzzled about the nature of the Cave Sandstone ; it 

 was so strikingly different from ordinary sedimentary beds, and had 

 the appearance of a trachytic tuff ; as all the lavas, however, round 

 about were of a decided basic type, I was at a loss to account for 

 acid material. An analysis by Mr. J. Lewis gave 83*5 of silica. A 

 microscopical examination showed besides the quartz grains, micro- 

 cline and plagioclase felspar, zircon, rutile, tourmaline, chlorite, 

 garnet, and epidote. The chalky look of the rock was produced by 

 a growth of white alteration products round the quartz grains.'" I 

 was obliged, therefore, to consider the white rock plugging the 

 vents as the triturated material torn off from the throat of the canal 

 which must have gone so far below the present surface of the land 

 that it tapped the deep-seated rocks underlying the newer sedi- 

 mentaries. These old rocks crop out in Natal, and are composed of 

 granite and crystalline schists, precisely the kind to yield the 

 minerals found in the Cave Sandstone and its prolongations into 

 the vents. 



Had there been any land surface composed of granite and crystal- 

 line schists near at hand from which the materials could have been 

 derived by the ordinary course of denudation, I might have hesitated 

 on putting forward this view. As it is, there are undoubted sedi- 

 mentary beds of the Molteno series underlying the Cave Sandstone, 



* Ann. Rep. Geol. Comm., 1902, Cape Town, 1903. 



