192 THE OUED GHAGHAIA TORRENT. ch. vm. 



On the summit of the ridge, which may be about 

 4,500 feet above the sea, the rock is a grey schist, often 

 shaly in texture, with the strike about east and west, and 

 dipping at a high angle approaching the vertical. These 

 beds may perhaps be identical with the schists, some- 

 times containing mica, and sometimes more calcareous 

 in composition, which we afterwards found at the head 

 of the Amsmiz valley, and with the rock, described 

 as micaceous schist, seen by Washington in his ascent 

 from Tasseremout. Our course now lay about due south, 

 parallel to that of the torrent which ran at a consider- 

 able depth below us. At Moulai Ibrahim this, accord- 

 ing to M. Balansa, is called Oued Ghaghaia, but we never 

 heard any similar designation for it. The difficulty of 

 seizing the shades of more or less guttural sounds from 

 the mouths of the natives makes it not improbable that 

 the word Grhaghaia of M. Balansa is the same that we 

 agreed in writing Eeraya, and that the name may mean 

 that this is the stream draining the district of Eeraya. 



On this ridge we found that curious grass, Lygeum 

 Spartum, characteristic of Sicily and Southern Spain, 

 where it is much used for making fine basket-work, but 

 not seen elsewhere in Marocco. Soon after we lit upon a 

 single specimen of a very fine plant of the artichoke 

 family, evidently distinct from all those described, but 

 unfortunately not yet in flower. It has been provisionally 

 named Gynara Hystryx (Ball). The next find was not less 

 interesting — an Oriental Echinospermum {E. barbatum 

 of Lehmann) that extends from the Punjab to Asia Minor 

 and the Caucasus, but had not before been seen in Africa. 



About two o'clock we left behind us the rough irre- 

 gular ground over which we had been riding, and found 

 ourselves in a broad open valley, with a level floor, half a 

 mile or more in width, at the head of which rose some 

 fine snow-seamed peaks. As we advanced towards the 

 main chain, our suspicion that the dividing ridge and the 

 higher peaks were at once more distant and more lofty 



