16 BARBOUR: ZOOGEOGRAPHY. 



Two species of Nectes, N. pleurotaenia van Kampen, and N. werneri van 

 Kampen, are confined to Sumatra and Borneo. 



To sum up, then, we may say that the portion of the fauna of Sumatra 

 under discussion shows nothing but a direct Malayan derivation, unless we 

 except Dyscophina, whose ancestors may have died out, or are as yet undis- 

 covered in Malacca. It has already been noted that those small islands which 

 lie in the Indian Ocean off the coast of Sumatra, whose fauna has been 

 derived directly from Sumatra, have not each of them the same common Su- 

 matran species; and Werner has remarked that this is probably due to the fact 

 that they have received their species by transportation from the nearest adja- 

 cent part of the larger island; and that one island may have received one typical 

 Sumatran species, and another island another, simply because these species were 

 confined to the particular parts of Sumatra which lay opposite the various 

 islands. This I hold to be improbable. He goes on to note how ridiculous it is 

 simply to record Sumatra without further data in keeping locality records of 

 species collected on the island; for he assumes, with excellent reason, that 

 probably many of the characteristic species are closely confined to certain areas 

 on the island. Concerning this we have as yet very hmited data; enough, 

 however, to prove that it is probable. 



Van Kampen has remarked, in a paper on the ampliibians of Sumatra 

 . (Zool. jahrb. Syst., 1905, 22, p. 714-15), that there is no evidence whatever to 

 be drawn from the distribution of Amphibia to show that there has been a con- 

 nection between Java and farther India — through the Mentawei and N icobar 

 Islands — without connection with Sumatra. It may be remarked that the 

 reptiles also bear out this statement; furthermore, as will be shown in the re- 

 marks on Java, Sumatra seems to have pro^'ided that island with the larger share 

 of its Malayan forms — with the larger share of all its forms, in fact. There 

 may have been a bridge of short duration between Borneo and Sumatra by way 

 of the islands of Billeton and Banka, for it is to be noticed, that if the main 

 mountain chain of Borneo were projected towards the southwest, it would reach 

 and merge with the low mountains existing on these two islands, which lie 

 directly between Borneo and Siunatra. The evidence, however, so far as the 

 reptiles and amphibians go, is very far from being convincing. 



NiAS. 



Closely related to the problems connected with the herpetology of Sumatra, 

 are those of the islands lying near by in the Indian Ocean. We have data re- 



