PHYSICAL GEOORAPHY OF ABYSSINIA. 155 



necessary to have produced such an effect must have 

 b'een very great. 



If the action of such small streams as the Jitta and 

 Bashilo has sufficed to sweep away the contents of 

 ravines 3,000 or 4,000 feet in depth, what may not have 

 been the effect of rivers like the Takkazzyt^ and Mareb ? 

 How much of the Abyssinian highlands has been removed 

 by these great torrents, and spread as an alluvial deposit 

 over the basin of the Nile ? 



There is every probability that the trap hills of Adi- 

 grat, described in a subsequent page, are a mere outlier 

 of the great mass of bedded trachytes and dolerites of 

 Lasta, Amhara, and Shoa. It appears probable, from 

 the accounts given by travellers, that the high ranges of 

 Samyen or Simen, south-east of Takkazzyd, are another 

 similar outlier. If this be so, then over a portion, pro- 

 bably over the whole, of Northern Abyssinia, there 

 existed at least 4,000 feet of bedded traps, of which now 

 only a few vestiges remain. 



It is worthy of repetition that throughout this great 

 denuded , area, so far as it was possible to examine it, 

 there is not a trace of marine denudation. The prevail- 

 ing features of the country are deep ravines cut by 

 rivers, and terraced hill-sides, moulded by the sub- aerial 

 disintegration of the rocks of which they are composed. 

 On all the slopes there are unequivocal marks of the 

 unequal action of surface weathering upon rocks of dif- 

 ferent chemical constitution. This is especially seen xipon 

 the traps. 



How far the great scarp of the Abyssinian plateau 



