162 BARBOUR: ZOOGEOGRAPHY. 



peculiar in that their distribution shows no- special correlation to that of the 

 reptiles. To be sure, they have been derived from the same two directions; and 

 the engystomatids from the Asiatic continent, which appear to have come from 

 Borneo through the southern PhiUppines, and so to Papua, met there the hylids, 

 and the single cystignathid, Aphaniotis novae-guineae van Kampen, which have 

 come from South America by way of Australia. That these amphibians haxe 

 been estabUshed in New Guinea for a very long time, perhaps even since the 

 beginning of the Tertiary, is shown by the great number of generic types which 

 are autogenous in Papua. 



With the reptiles of New Guinea we find that specific differentiation is 

 general, but that only a few genera are entirely confined to the island. This 

 suggests at once that the amphibians reached New Guinea long before the 

 reptiles, and this is probably the case; though there seems no special reason to 

 assume that the later migration of reptiles came along a different highway than 

 that over which the ancestors of these peculiar endemic genera of amphibians 

 had already passed. It is not necessary to assume that both reptiles and amphi- 

 bians had coincident periods of maximum activity of dispersal, since there is no 

 reason to believe that the amphibians are much more plastic, or much more 

 subject to evolution by isolation, than are reptiles. This condition of these two 

 groups in Papua is in marked contrast to the condition in Sumatra, for instance, 

 an island which was long an actual part of the continent, and not a distant area 

 connected by narrow and perhaps short-lived bridges with the most outlying 

 region of the continental area. It seems fair to state that the amphibians of 

 New Guinea form a very old fauna, derived even before the migration of opistho- 

 gljrphs into Australia, and certainly of much older origin than the greater part 

 of the reptilian fauna of New Guinea itself. 



The finding of Cornufer widespread from Fiji to the Philippines, as well as 

 the development of Ranas and of Rana derivatives in the Solomon Islands, 

 would make it appear probable that the Ranidae came to Papuasia before the 

 Engystomatidae, or else that the latter, in spite of their adaptation to xerophilous 

 life, spread less successfully. 



Another explanation of this state of affairs which may be a solution to the 

 question is that the Ranidae may have come from Celebes to the southern Mo- 

 luccas, and so to Papua, and passed on quickly to the Solomon Islands and Fiji; 

 while the Engystomatidae may have come through the southern Philippines to Hal- 

 mahera, as has been indicated before, and, spreading more slowly, found the Bis- 

 marck, Solomon, and Fiji Archipelagos cut off from Papua before their advent. 



