306 MR. «. A. ROLFE ON THE 



history." He also says (p. 346) : — " We find apparently two sets 

 of animals — a more remote series ... in which the species are 

 distinct from any others, and a more recent series ... identical 

 with common Malayan animals. The former indicate the earliest 

 period when these volcanic islands were connected with some 

 part of the Malayan subregion. . . . The latter may indicate either 

 the termination of the period of union or merely the effects 

 of introduction by man. The reason why a larger number of 

 mammalian forms were not introduced and established was 

 probably because the union was effected only with some small 

 islands, and from these communicated to other parts of the 

 Archipelago ; or it may well be that later subsidences extin- 

 guished some of the forms that had established themselves." 

 Four years later (' Island Life,' p. 301) he gives a somewhat 

 different explanation of the peculiar phenomena; he says : — "it 

 is evident that the Philippines once formed part of the great 

 Malayan extension of Asia, but that they were separated con- 

 siderably earlier than Java. . . . The reason of their comparative 

 poverty in genera and species of the higher animals is, that they 

 have been subjected to a great amount of submersion in recent 

 times." 



If the Philippines had once formed part of the great Malayan 

 extension of Asia, which was separated earlier than Java, it 

 would have required an almost total submersion to have caused 

 such wholesale extinction of typical Malayan tribes and genera ; 

 while, on the other hand, many of the survivors are not more 

 fitted, and some much less so, to survive such a submersion, than 

 many of those believed to be absent at the present time. Geo- 

 logical evidence will probably in future throw much light on this 

 point. Meantime the present evidence appears to support the 

 former of these two opinions. It has been shown that nearly all 

 the genera have their headquarters elsewhere, the few which are 

 endemic being principally mouotypic and closely allied to those 

 of surrounding countries ; thus proving that the present flora 

 has been chiefly or altogether derived from these sources, and is 

 consequently of more recent origin, and that in many cases the 

 lines of migration can easily be traced. It has also been shown 

 that, with the exception of a large number of species of very wide 

 distribution, or that have been directly or indirectly introduced 

 by human agency, this migration (which occurred after a period 

 of stupendous volcanic upheaval) was sufficiently remote to 



