MARCH FROM SENAFE TOWARDS MAQDALA. 55 



expectations that a portion at least of the army would 

 remain in Abyssinia throughout the monsoon were shared 

 by a large majority, if not by the whole, of the expedi- 

 tionary force. As will be seen, I left just in time to 

 catch the force before Magdala. 



Leaving Senafe, the road traverses a plain of meta- 

 morphic rocks, excessively slaty for the most part, and 

 showing very little crystalline structure. To the west 

 is a range of hills capped by sandstone ; to the east 

 the slaty and schistose formations form rounded masses, 

 between which flow the small streams which form the 

 heads of the valleys running towards the Salt Plain near 

 Amphila. There is but little of interest on the road until 

 it approaches Guna Guna, twelve miles from Senaf^. 

 Here the sandstone scarp to the west approaches nearer 

 to the road, and in a small cleft in it is the shrine of 

 St. Eomanos, where lie his bones and those of his com- 

 panions. The spot is a very lovely one : a little stream 

 runs through a ravine with precipitous sides, the bottom 

 fiUed with magnificent trees. A little farther, the road 

 enters the valley of Guna Guna, with its flat bed of 

 rich turf surrounded by magnificent sandstone precipices, 

 half-way up one of which is a church cut in the 

 rock. 



I stayed here a day examining the sandstones, which 

 are precisely similar to those of Sowera, and in which I 

 could find no traces of fossils. I shot a sparrow-hawk, 

 which I subsequently found to be a fine old specimen 

 of the rare Accipiter unduUventer of Euppell, and 

 I also, for the first time, saw that remarkable African 



