242 EAST AFRICAN FLORA AND FAUNA part in 



present homes of the species. In Europe this admits of very 

 easy explanation, for the former extension of the glaciers in the 

 great Ice age drove the northern plants farther south ; as the 

 climate became milder and the ice receded, the plants returned 

 northward, and in the south survived only on the summits of 

 the higher mountains. Analogous cases have been found in 

 North America, and admit of the same explanation. Thus on 

 the White Mountains in New Hampshire there is a group of 

 butterflies of Arctic types, which have been left as a legacy 

 from glacial times. 



The occurrence of these isolated patches of an Alpine flora 

 in Equatorial Africa therefore suggests that, in times geologi- 

 cally recent, a change has come over the climate. This is con- 

 firmed by other lines of evidence ; instances are known of the 

 survival of plants in situations now ill-suited to them. For 

 example, at the southern end of Basso Narok, von Hohnel 

 found a patch of the Hyphaene palm. His photograph shows 

 that the trees are dwarfed and unbranched, and are clearly 

 growing under unfavourable conditions. They are probably 

 survivals from a rich growth of these palms that flourished 

 around Basso Narok at a period when a great river flowed into 

 its southern end, and its waters were less alkaline than they 

 are to-day. The geographical evidence is equally conclusive. 

 Livingstone ^ has graphically described the existence of river 

 gorges in Bechuanaland far larger than could have been cut 

 by the rivers that now flow through them ; and the same fact 

 has been reported in the Sahara by Zittel, Rohlfs, and Weld 

 Blundell. Old beaches and terraces indicate the former exist- 

 ence of lakes, whose waters have long since been lost by 

 evaporation into the air and absorption by the soil. 



Hence the evidence of the plants, river gorges, and old lake 

 basins, together demonstrate that some change has taken place 

 in the African climate. The two latter could be explained by 

 a mere decrease in the rainfall, but this would not account for 

 the first ; and the facts afforded by the Cape, Kilima Njaro, 

 and the Andes seemed to forbid an appeal to the glacial 

 agencies, which explain the analogous cases in Europe and 

 America. 



^ D. Livingstone, Missionary Travels arid Researches in South Africa (1857), 

 p. III. 



