Vol. 50.] OF SOUTH-EASTERN AFRICA. 551 



I do not intend to enter into a minute description of the rocks 

 composing the various groups in South-eastern Africa, but merely 

 wish to place on record their mode of occurrence and the position 

 they occupy. 



(1) Volcanic Beds. 

 ( Trap-amygdaloids. ) 



These occur only on the tops of the higher peaks of the Drakens- 

 berg, in the mountain-portion a (see p. 548). They form a most 

 fantastic capping to the mountain-range, rising in peaks and pin- 

 nacles to 1000 feet in height above the sedimentary rocks of the 

 range. 



This group does not extend farther northward than the Mont- 

 aux-Sources, and is rarely found lower than 8000 feet above sea-level. 

 None of the higher eminences of the plateau or High-veld portion of 

 the Drakensberg are capped with these rocks, which apparently 

 never covered the sedimentary deposits northward of the Mont-aux- 

 Sources. The Cave-sandstone forms the hill-tops of the plateau b. 

 (PL XXII.) 



(2) Cave-sandstone. 



This group attains a thickness of between 500 and 700 feet, and is 

 found underlying the volcanic beds, wherever they occur on the 

 Drakensberg range. 



It forms the crag-crowned tops of most of the hills situated on 

 the plateau or 6-portion of the Drakensberg, but does not extend 

 northward of the Vaal River on to the High-veld plateau. If it ever 

 was deposited there, it has been completely removed by denudation. 



The specimens of Semionotus capensis, Cleithrolejois Ecctoni, and 

 a new species of Dietyopyge (?) described by Mr. A. Smith Wood- 

 ward (see Ann. & Mag. Nat. Hist. ser. 6, vol. xii. 1893, p. 398, 

 pi. xvii. fig. 1), have been found in the Cave-sandstone in the 

 neighbourhood of the villages of Ficksburg and Rouxville, in the 

 Orange Free State. 



(3) Red Beds. 



This group occupies a small area in the vicinity of the town of 

 Harrismith (O.F.S.), and is exposed in the lower portion of the Plat- 

 Berg, on the town-commonage, extending southward into ' Vitsies ' 

 Hoek, towards the Mont-aux-Sources. 



It is about 100 feet thick at the section near Harrismith, and 

 contains a bed of bone-breccia, principally composed of reptilian 

 remains. Specimens from this bed were sent to the Royal College 

 of Surgeons by Messrs. Orpen, and noticed by Prof. Huxley in the 

 Society's Quarterly Journal, vol. xxiii. (1867) p. 5, having been 

 previously described by Prof. Owen in his Catalogue of the Fossil 

 Reptilia in the Museum of that College (1854). 



Northward, towards the Vaal River, the Red Beds do not appear 

 to exist, but seem to be replaced by a dark-coloured, gritty sandstone, 

 containing worn crystals of felspar, of considerabl 3 sizp and rounded 



