ZOOGEOGRAPHY. 163 



Mention has been made elsewhere in this paper that, broadly speaking, 

 the number of species supported by any given island bears a direct relation to 

 the area of the island in question. This is not an easy theory to confirm with 

 definite data, inasmuch as two islands that appear to be geographically of about 

 the same size may in reality be vastly different in superficial areas, according 

 to the surface features of the country. Thus, in the West Indies, Jamaica and 

 Porto Rico appear to be of about the same size. It is evident at once, however, 

 that the fauna of Jamaica is vastly greater than that of the other island. It may 

 be argued that this is due to Jamaica's being nearer to the mainland source of 

 supply; but the fact that the whole island is traversed by high mountain ranges 

 of broken peaks, while Porto Rico has only one or two isolated highland areas, 

 would seem to offer the true solution of the question. So in the East Indies, 

 taking the islands in the order of their size, the numbers of species of reptiles 

 and amphibians are about as follows: — 





Amphibia 



Reptil 



New Guinea 



70 



144 



Borneo 



79 



207 



Sumatra 



48 



166 



Celebes 



26 



83 



Java 



37 



122 



Papua and Borneo are of practically the same size. Recent accounts of 

 surveys of them indicate that New Guinea is really the larger island. Enormous 

 ranges of high mountains give it, of course, a vastly greater surface area. At the 

 rate at which new species have been found in New Guinea, during the last few 

 years, it will soon take its place at the head of the list. In Borneo, on the other 

 hand, new species of reptiles and amphibians are now but rarely found, and our 

 systematic knowledge of the fauna is probably approximately complete. Celebes 

 has a larger area than Java; but, on the other hand, it has many fewer peaks, its 

 highlands being mostly of the nature of plateaux, while in Java there are many 

 chains composed of high peaks rising from lowland areas. ' This, coupled with 

 the fact that it is somewhat nearer the source of supply whence the fauna of 

 Celebes has come, gives it a larger fauna than seems to be present in Celebes; 

 though here, again, we must remember that Java is the best known island in the 

 entire archipelago. The fact that the Ke Islands, which, zoologically, seem to be 

 very closely related to Ceram, have so few amphibians and so many less reptiles 

 than Ceram is probably due to the same factor. When the Moluccas are really 

 well explored, we shall probably find that Halmahera holds the lead it now has in 

 point of number of species found on it; while Ceram will come next, and Buru, 



^^ii 



