Dr. J. W. Gregory — Geology of Somali-land. 293 



Whether it has been separated by denudation or earth-movements, 

 there is not at present sufficient evidence to show. 



The Somali plateau itself consists in the main of a mass of gneiss 

 and rocks of the Archean series. As usual, this gives rise to the 

 sandy scrub -covered plains, with many mimosa and Sanseviera. 

 The Archean series is penetrated by pegmatite dykes of the same 

 type, but coarser in grain than those of British East Africa. It is 

 capped by rocks of two series — (1) red and purple unfossiliferous 

 sandstones, often passing into impure haematite ; (2) some limestones 

 associated with beds of chert, prohably of Neocomian age. These 

 form the cap of the plateau aud the summits of the Golis Range. 

 The ledge of Mirsa and the summit of the Sheikh Pass are both 

 on the upper surface of the Archean series (4000 feet). 



The Somali-land sequence therefore consists of the following: — 



1. Marine and subaerial recent deposits. 



2. Neocomian limestones and cherts. 



3. Duba limestone of uncertain age. 



4. Bathonian limestones of Bihin. 



5. Red and purple sandstones — unfossiliferous. 



6. Archean series. (Gneisses penetrated by pegmatite dykes, etc.) 



V. The Origin of the Somali-land Fauna and Flora. 



The discovery of the Bathonian fossils is of great interest. 

 Bathonian rocks are known in Ahyssinia from the work of 

 Blanford, 1 Ferret and Galinier, 2 Aubry and Douville. 3 They are 

 also well known from the west coast of India in (Dutch. They 

 occur in Northern Somali-land, exactly on the line joining these 

 two localities. Mrs. Lort-Phillips' fossils are especially of value, 

 as Rochebrune, 4 on the basis of Revoil's collections from Somali- 

 land, has disputed the accuracy of the determination of the Jurassic 

 age of the Abyssinian limestones. The presence of Bathonian fossils 

 in Somali-land, however, shows that, apart from the Abyssinian 

 limestones, there are fossiliferous marine deposits of both Lower 

 Jurassic and Neocomian age in this region of East Africa. 



The occurrence of one species (Parallelodon Egertonianus, 

 Stoliczka) in Somali-land and in India helps to strengthen the 

 idea of the former connection of the Bathonian rocks of India 

 and Somali-land, which is suggested by the apparent extension of 

 these rocks along an east and west line, from Cutch to Shoa. 

 This is of interest, as it throws light on the origin of the 

 interesting fauna and flora of the Golis Range. The report on the 

 collection of plants made in the Golis Range, published in the Kew 

 Bulletin of Miscellaneous Information (No. 105, September 1895, 

 pp. 211-30), shows that the flora of this district contains a very 



1 W. T. Blanford, "Observations on the Geology and Zoology of Abyssinia," 

 1870, pp. 176-80, 199-203. pi. viii. 



2 Ferret and Galinier, " Description geologique du Tigre et du Samen Voyage en 

 Abyssinie," t. iii, 1847, pp. 54-6. 



3 H. Douville, " Examen des fossiles rapportes du Choa par M. Aubry" : Bull. 

 Soc. geol. France, ser. 3, t. xiv, 1886, pp. 223-41, pi. xii. 



4 Rochebrune, in Bevoil, op. cit., p. 26. 



