522 PROFESSOR J. W. GREGORY. 



African coast, as in the French Congo near the mouth of the Gaboon. The Angola 

 Cretaceous beds were probably deposited in an arm of the Tethys, which ran down 

 the western coast of Africa. The evidence for a connection with India around South 

 Africa appears as yet inadequate. 



A lower Mesozoic horizon may be represented by the gypsiferous sandstones with 

 wind-rounded grains from the head of the Lengwe Gorge. The sandstone of Dombe 

 has been repeatedly referred to the Trias (see, e.g., Choffat, 1888, p. 4l). # Choffat 

 (1888, p. 45, and 1905, p. 56) suggests that it is the same age as the coal-bearing 

 sandstone of Dondo and of the Dande valley. These three localities are in the 

 north-western corner of Angola, and Choffat (1888, p. 45) considers that this 

 horizon is also represented further south at Novo Redondo and at Dombe Grande. 

 The sandstone at the head of the Lengwe Gorge probably belongs to the same group, 

 as its lithological characters indicate that it was laid down under different climatic 

 conditions from those of the Cretaceous, and it is not improbably of Lower Mesozoic 

 age. Some of the deposits at the base of the Cretaceous Series, such as the gyp- 

 siferous beds resting on the gneiss at Hanha or the sandstones or conglomerates at 

 the foot of the Lengwe Gorge, are perhaps also pre-Cretaceous. 



B. The Older Sedimentary Rocks on the Plateau. 



Upon the ancient gneisses which form the foundation of the Benguella plateau 

 rest disconnected areas of sedimentary rocks, which have hitherto yielded no fossils. 

 Their correlation may therefore appear impracticable. The wide distribution of old 

 unfossiliferous sediments is one of the most striking features in the geology of 

 South and Central Africa, and fortunately the geologists who have worked in 

 those regions have had the courage to attempt the correlation of these deposits by 

 their lithological characteristics and the extent to which they have undergone dis- 

 location. These rocks clearly belong to various ages, and it is recognised that they 

 range from the pre-Cambrian to the Carboniferous. 



The older sedimentary rocks which I saw on the Benguella plateau (excluding 

 those near Lengwe) may be divided into five groups — 



(1) The Bihe Sandstone. 



(2) The Oendolongo Series. 



(3) The Lepi Graywackes. 



(4) The Huambo Quartzites. 



(5) The Bailundo Schistose Quartzites. 



In the absence of paloeontological evidence the correlation of these rocks depends 

 on their resemblances in structure and succession to those in South and Central 

 Africa. It is natural to compare them first with the long and varied succession of 



* Mr A. Holmes (1915, p. 231) also refers to the Dondo coal-bearing shales and grits of North-Western Angola 

 as of Lower Karroo age. (,'oknet (1894, p. 195) says it is pre-Jurassic. 



