GEOGRAPHICAL DISTRIBUTION 143 



cussion of geographical distribution requires a clear under- 

 standing of certain points which may be briefly alluded 

 to here. The situation with plants is, of course, quite 

 different than that with animals. With the advance of a 

 continental ice sheet, for example, animals may actively 

 retreat, by their own locomotion. There are exceptions 

 to this method of animal dispersal. Insects and small 

 birds may be blown by the wind over considerable dis- 

 tances, and insect eggs, larvae, and cocoons may be trans- 

 ported in the soil about the roots of floating uprooted trees 

 and otherwise; and instances are on record of animals 

 being carried as passengers on floating objects, notably, 

 according to Semon, in the Malay Archipelago; snakes 

 and crocodiles are known to have drifted in this way to 

 the shores of the Cocos or Keeling Islands, a group of 

 coral atolls in tlj.e Indian Ocean, about 700 miles south- 

 west of Java, the nearest land. 



But plants, at all times and under all circumstances, 

 are wholly dependent on being carried passively by exter- 

 nal agencies. The chief means of seed dispersal are the 

 wind, streams and ocean currents, and animals particularly 

 birds. For distribution over great distances it is of im- 

 portance to ^consider^chiefly wind, ocean currents, and 

 birds. 1 



114. Dispersal by ^ Wind. DeCandolle, the great 

 Swiss student of plant geography, regarded the wind as 

 "the most general and ordinary cause of the distribution 

 of species over the entire surface of a country," but re- 

 jected it as a means of dispersal over even narrow arms of 



1 The distribution of seeds in connection with commercial shipments is 

 interesting, but not essential to our present purpose. The word "seeds" 

 is used above to designate all reproductive bodies, including fruits, spores, 

 and vegetative reproductive bodies, such as gemmae, bulbils, etc. 



