2i6 THE GEOLOGY OF BRITISH EAST AFRICA part hi 



return journey, we traversed it farther to the north. As this 

 plain was therefore lava and not alluvium, it took its place as 

 part of the great volcanic series of Africa, the approximate extent 

 of which is marked on the accompanying sketch-map (Fig. 7). 

 An inset map of Southern Italy on the same scale shows the 

 insignificance of the lava flows of Vesuvius and Etna in com- 

 parison with those of Eastern Africa. The latter cover so 

 vast an area, that it is useless to compare them with the flows 

 from ordinary volcanic vents ; they must be due to the same 

 method of eruption as that which discharged the rocks of the 

 great lava sheets of America and the Deccan Traps in India. 

 The extent of these is equally enormous ; the latter cover 

 an area of 200,000 square miles, and those of the Western 

 States of America have been estimated to occupy a tract of 

 country as large as Great Britain and France combined. More- 

 over, these vast lava fields have been described as free from the 

 tuff and ashes which form the larger part of the material 

 ejected by volcanoes. Hence, for two reasons, it seemed hope- 

 less to explain their formation by the ordinary type of volcanic 

 action. Baron von Richthofen therefore proposed the theory, 

 which was subsequently supported by Sir Archibald Geikie, 

 that these lava seas were discharged from subterranean reser- 

 voirs through fissures, possibly hundreds of miles in length, 

 instead of through simple circular vents. That cracks may 

 form for great distances across a country and be filled by 

 igneous materials is proved by the occurrence of dykes, 

 which cut across northern England and Scotland ; that the 

 contents of such fissures occasionally reach the surface is known 

 from observations in the Sandwich Islands. It was obvious at 

 once that these Kapte plains were older than those of North 

 America, and there was accordingly more chance of finding 

 exposed upon them some of the supposed fissures. A ride on 

 the great lava sheets of the Snake River of Idaho in 1891 had 

 temporarily shaken my faith in the fissure-eruption hypothesis, 

 but it had not destroyed it, and I therefore left the Iveti 

 Mountains for the plains expecting these would yield im- 

 portant evidence in support of the theory. The rocks have 

 been extensively denuded ; the Athi has cut a deep gorge 

 through the plain, but nowhere could I find any sign of the 

 fissures. In places many small dykes occur, but their existence 



