VIII.] SUB-REGIONS. 307 



species with Malayan forms and the absence of any peculiar 

 genera save Phlceomys. The absence of such a number of 

 Malayan types indicates, however, either that the connection must 

 have been exceedingly brief, or that a large number of species 

 formerly inhabiting the islands have been destroyed by sub- 

 mergence. On the other hand, the presence in the group of a 

 considerable proportion of widely spread continental genera of 

 birds suggests a more free communication with China than at 

 present exists ; such communication having not improbably taken 

 place by way of Formosa. The mountain fauna of Luzon doubt- 

 less indicates an earlier type of colonisation. 



Note. Since the foregoing was written two papers have appeared on the 

 Fauna of the Natuna Islands; viz. O. Thomas, Novitates Zool. Vol. n. pp. 

 2628, and Thomas and Hartert, I.e. pp. 409 429. 



2O 2 



