THE DWYKA FORMATION 407 



others at Laingsburg and went by train via Johannesburg to Vryheit 

 (VY, figure 1) in Natal, to join a party led by Mr W. Anderson, geolo- 

 gist of that colony, and Dr G. A. F. Molengraaff, formerly geologist to the 

 South African Eepublic. From Vryheit we drove some 30 miles north- 

 ward, past a dolerite-capped table mountain called Hloban (HI, like the 

 Welsh LI), where General Buller some twenty years ago defeated the 

 Zulus with the aid of the Boers and won the Victoria cross; past the 

 ruined farm of General Botha, around which the long avenues of Euca- 

 lyptus still flourish, to a small inn on one of the Emmet farms, called 

 Waterval. The next day we drove a dozen miles farther to a valley on 

 the headwaters of Pongola river, which flows down the eastward slope 

 where the interior highland descends toward the Indian ocean. There, 

 on August 24, we saw the horizontal Dwyka-Ecca series, 2,000 or 3,000 

 feet thick, rising in a dolerite-capped table mountain or plateau remnant, 

 called Ngotshe (N, figure 1; the name begins with an unwritable Zulu 

 "click"), and resting unconformably on a floor of Barberton beds, hard 

 slates of early Paleozoic or older date, which dipped 24 degrees to the 

 northwest. This district was explored by Molengraaff in 1897, when he 

 found several highly significant contacts of the older and younger forma- 

 tions in certain dry ravines on the north slope of the Ngotshe mass; it 

 was one of these contact localities that we came to see. 



The Dwyka is here closely packed on an uneven surface of Barberton 

 beds, which are beautifully rounded, grooved, and striated — in a word, 

 well glaciated. Sometimes a ledge of the scoured and striated surface 

 overhangs by a small amount; the Dwyka is then packed in beneath such 

 a projection. The ravine that we followed gave excellent sections of the 

 tillite, as in plate 51, figure 2, and often exposed the underlying Bar- 

 berton surface with more or less distinct grooving and striation, and with 

 more or less Dwyka tillite (plate 53, figures 1 and 2) resting on it, for a 

 quarter of a mile or more. The grooves and striation s on the Barberton 

 hold their courses unchanged as they pass under the tillite cover, remind- 

 ing one of the manner in which the wheel tracks in the Eoman forum 

 pass straight under the cover of detritus at the side of the modern excava- 

 tion. The relief of the Barberton surface seemed to be of moderate 

 measure in this immediate locality, probably not more than 20 or 30 feet 

 in the dry ravine that we followed; but the relief of the present surface 

 is to be measured in thousands of feet, from the dolerite-capped rem- 

 nants of the plateau of the horizontal Karroo series to the bottoms of the 

 larger valleys, deep cut in the ancient Barberton slates. 



The Dwyka itself repeated all the features previously noted. A finely 

 striated 5-foot boulder, with Penck's half-meter hammer handle lying 



