CHAP. XIII FORMER RANGE OF ALPINE FLORA 245 



possession of narrow, spiny, or needle-shaped leaves and succulent 

 leafless stems would not then have been an advantage ; the 

 foliage would have been more luxuriant and better adapted for 

 animal food ; forests that now occur only as belts beside the 

 rivers would have been spread far and wide across the country. 

 The scrub would have been replaced by woodland, and districts 

 that now, as barren sandy deserts, present barriers to animal 

 migration, would have been fertile, well-watered prairie. All 

 the conditions, in fact, that govern the distribution of animal 

 and plant life would have been different from those that obtain 

 at the present day. 



At the time, therefore, of the maximum size of the Kenyan 

 glaciers, the Alpine flora would have spread at least throughout 

 the areas marked by dots on the accompanying map (Fig. 13). 

 No doubt it extended lower, as traces of it have been found in 

 the mountains of Usambara, but the precise lower limit cannot 

 be determined from the data available at present. When the 

 clima\:e changed with the decrease of the glaciers, the Alpine 

 flora crept up the mountain sides, and became isolated from its 

 allies in the temperate regions to north and south. 



We have assumed, however, that this Alpine flora has 

 received contributions not only from the north but from the 

 south, and this compels us to consider the theory that 

 all faunas and floras have originated in the northern hemi- 

 sphere, and thence have worked their way to the south. The 

 occurrence of isolated patches of the flora of one region in 

 a more southern zone has been mentioned as not unknown in 

 Europe and North America. Similar evidence occurs in Asia. 

 On the mountains of Ceylon and the plateaux of Southern India 

 there occur plants such as Rhododendron arboj^eiuti, Sm., and 

 mammals such as the Indian Marten {Martcs Jlavigida, Bodd.), 

 which do not live on the surrounding lowlands, but reappear to 

 the north on the Himalaya. Similarly, the Himalayan fauna and 

 flora can be traced along the mountains of Assam and through 

 Burmah into the Malay Peninsula. An outlier occurs even on 

 the hills of Java.^ Russell Wallace^ and Haacke^ have 

 collected a good deal of evidence of the same character, and 



^ Blanford, Medlicott, and Oldham, Afatt. Geol. hidia, ed. 2 (1894), p. 14. 

 ^ A. Russell Wallace, Island Life (1880), chap, xxiii. 



3 W. Haacke, " Der Nordpol als Schopfungszentrum der Landfauna," ^/t?/. Centralbl. 

 Bd. vi. (1886), pp. 363-370. 



