376 SCIENCE PROGRESS. 



is marked in the north-western part of the Sahara. The 

 whole of the Ioidi Desert and most of the land between this 

 and the Atlas is included in the belt. But the country of 

 Tiris, which lies north of Cape Blanco, is coloured as a 

 wedge of Quaternary deposits between the older rocks and 

 the sea. Vincent, however, long ago described Tiris as a 

 granitic plateau (24, p. 23), and more recently Ouiroga has 

 shown that the greater part of the district is occupied by 

 granitic and gneissic rocks (25). Some of these he refers 

 to the Archaean, while others (including an elaeolite syenite.) 

 he believes to be eruptive. There is a narrow band which 

 he considers to be of Palaeozoic age. 



Douls, too, describes oranite on both sides of the 

 Wadi Chebika (south of the Wadi Draa) (26, p. 456) ; and 

 it is probable that granitic rocks form a belt along the west 

 coast, separated from the sea by a narrow band of Pliocene 

 and recent deposits. 



This belt of crystalline rocks, however, does not appear 

 to extend far to the south. West of Shinghit, Panet (27, 

 pp. 103, 104) describes the occurrence of slate, agglomerate 

 and sandstone, together with what he believed to be basalt. 

 Lenz has coloured these as Silurian. In Senegambia, too, 

 which Lenz himself has traversed, his map shows the whole 

 region to be made of slate and quartzite, with laterite in the 

 valleys (22). It is not till we reach the Kong Mountains 

 that we again come upon an extensive area of gneissic 

 rocks. 



The gneiss of these mountains probably extends into the 

 bend of the Niger nearly to Timbuktu ; for Barth, on his 

 journey from Say to that town, noted granite at several 

 points (see map in 18, vol. iv.). The most remarkable, 

 however, of the rocks which he observed, were in the 

 Hombori Mountains. In his sketch (18, iv., p. 336) these 

 hills appear like lofty towers resting upon conical mounds of 

 detritus. It is difficult to believe that gneiss or granite 

 could weather into these forms, and they are more sug- 

 gestive of columnar lavas. Barth himself does not state the 

 character of the rock, although on his map he records a 

 granitic peak close by. But whatever the Hombori hills 



