NO. 2263. NEW AFRICAN EARTHWORMS— SMITH AND GREEN. 155 



past by such a land area as has been assumed under the name 

 Gondwana, somewhere in that region from Haplotaxid ancestry a 

 Moniligastrid branch may be supposed to have arisen. This branch 

 maj^ have split early into a Dcsmo(/asfer branch which reached and 

 retained a hold in the Burmese and East Indian regions; and a 

 Syngenodriline branch which reached and retained a hold in the 

 tropical East African region. From the latter branch the more 

 primitive Acanthodriline forms may have been derived, and other 

 descendants with but slight change have given rise to our present- 

 day genus Syngenodriltis. The suggested relationships are shown 

 in the diagram (fig. 9). 



Haplotaxidae 

 Pelodrilus 



Moniligastrid ancestor 



Syngenodrilinae Moniligastiinae 



Syngenodrilus Desmogaster 



Acanthodrilinae Monilujaslcr, Drawida, 



Diplocardiinae Eupolygaster 



Fio.9.— Diagram SHOWING suggested relations or Syngenodrilus 



AND RELATED QROXn'S. 



Because of the lack of adequate knowledge of the earthworm 

 fauna of the territory lying between Northern India and British East 

 Africa, we are, of course, not certain that primitive Moniligastrid 

 forms may not exist in the intervening territory ; nor that migration 

 may not have taken place through this territory, or even possibly by 

 the route postulated by Michaelsen (1910 : 28), by which the ancestral 

 forms of Eudichogaster have been supposed to have passed between 

 Africa and India. 



The absence of primitive Moniligastrid species from the Hindu- 

 stan peninsula and their presence both east and west of that region 

 presents a problem of a sort often encountered in dealing with primi- 

 tive groups having discontinuous distribution. 



The adoption of the ideas of Matthew (1915) on the general prin- 

 ciples of geographical distribution will give a very different notion 

 of the sequence of events leading up to the present distribution of 

 earthworms from that held by some writers. These ideas are based 

 on an acceptance of theories of continental development involving 

 the principles of isostasy. They assume a general permanency of 

 the great ocean basins and of the continental shelves. They also 

 assume an alternation of periods of continental elevation, aridity, 



