ZOOGEOGRAPHY. 161 



through the southern Phihppines. Mindanao and Halmahera both support a 

 number of engystomatid genera; butthese forms are not abundant in Celebes, 

 nor, so far as we know, in the Lesser Sunda chain, where we would expect to find 

 more of them if they got to Papua by this route. The few occurring there 

 represent probably what is really a back-flow from Halmahera. Borneo is well 

 known herpetologically; but these small, inconspicuous frogs are very difficult 

 to collect, and many may yet remain undiscovered both here and in the southern 

 Philippines. Most of the Papuan species are known by only few examples, 

 and new forms are being discovered constantly. I have described the most 

 recently found generic type, Pomatops valvifera Barbour, from a single example 

 taken from the stomach of a Matrix mairii Gray. That the land-bridges by 

 which such forms as these got from Malaya to Papuasia were narrow necks of 

 land that probably rose steeply from the sea is suggested, as van Kampen has 

 remarked, by the fact that the great majority of the types which were successful 

 in extending their range to Papua are those which in their development skip the 

 free-swimming larval period, and escape from the egg as completely meta- 

 morphosed adults. This is doubtless an adaptation to life in a land where there 

 is little or no standing water, which was probably the condition on these narrow 

 stretches of land. We know certainly that there is no dearth of standing water 

 suitable for spawning places on either Borneo, Mindanao, or Papua. So far as 

 we know, Ceram is very poor in amphibians, as are also Ambon and Buru, 

 though of the latter we know practically nothing. So, for this reason again, 

 it seems likely that the southern Moluccas were not in the direct route of mi- 

 gration of the Engystomatidae, or of the Rana varians- or R. moluccana-like form 

 which gave rise to Rana papua in New Guinea. The connection of Ceram with 

 Papua may have been of so narrow a form and of so transitory a nature that 

 reptiles and birds passed across easily, but that only few amphibians came over. 

 Again the Ceram-Ke-Papua Bridge, or the Ceram-Mysol-Papua Bridge may have 

 existed before so many varied amphibian types became differentiated from one 

 another by some process akin to what Osborn has called the law of adaptive 

 radiation; or the bridges between Ceram and Papua may have already disap- 

 peared before the influx of amphibians by way of Halmahera. This is quite 

 possible, since Mysol, which probably supports an amphibian fauna similar to 

 that of Papua (though we know nothing of it as yet), would have lain directly 

 in the migration route from Halmahera to Papua, and hence would have passed 

 on these amphibians to Ceram, if a suitable bridge had existed for them to cross 

 on. The amphibians now existing in New Guinea strike one at once as being 



