IX] GEOGRAPHICAL DISTRIBUTION 139 



Although the genus Notiodrllus certainly, and 

 Microscolex possibly, extend into the tropical regions 

 of America, Africa, and Australia, these species are 

 but few, and the bulk of the species and of the allied 

 genus Chilota are restricted to the antarctic quarter 

 of the globe ; they also extend all over it, that is to 

 say in the southernmost parts of South America, 

 in the Cape region of Africa, in Kerguelen and the 

 Crozet Islands, and in New Zealand, as well as in the 

 Auckland Islands and other neighbouring islands. 

 It is true that I have excluded New Zealand from this 

 region on the grounds that it forms a debateable 

 ground between it and the Indo-Australian. But 

 apart from this part of the world the rest of the 

 territories mentioned should be combined to form 

 the antarctic region. 



Having therefore arrived at a mapping out of the 

 world into regions in accord with its earthworm 

 fauna, it is desirable to ascertain what light the facts 

 throw upon the geological and evolutionary questions 

 with which the study of zoogeography deals. The 

 existence of an antarctic region binding together such 

 distant points as South Georgia, the Cape of Good 

 Hope and Kerguelen Island, seems to argue strongly 

 for the former extensions northwards of the antarctic 

 continent so far north as to embrace these several 

 regions of that hemisphere. In view of the facts 

 relating to the danger of sea water to earthworms, to 



