434: TRANSACTIONS OF SECTION D. 



Group in the Oriental Region, substituting ' Weber's Line ' ' for ' Wallace's 

 Line ' as the zoological boundary between Asia and Australia. In the opinion 

 of some authorities Wallace's Line has scarcely any importance ; thus a con- 

 tributor to a recent edition of the ' Encyclopaedia Britannica ' writes of ' the 

 now exploded Wallace's Line,' and Professor Max Weber calls it 'this unfor- 

 tunate line ' ! He has written at considerable length on the origin of the fresh- 

 water fish fauna of the Malay Archipelago, 2 and holds that that of Celebes has 

 no Australian, but a highly impoverished Indian character, and that there is 

 no sharp boundary between Bali and Lombok, the southern impoverishment 

 beginning in Java and becoming more marked in Bali. 



I think that to illustrate the geographical distribution of fresh-water fishes 

 it is necessary to make a primary division between the Australian Region, in- 

 cluding the islands to the east of Wallace's Line, and the rest of the world. 

 In the Australian Region we have two archaic types (Ceratodus, Scleropages), 

 but all the other fresh-water fishes belong to marine families or genera. 

 Peculiar genera of sea-perchase (Serranidce, Kuhliida), sand-smelts (Atherindce), 

 gobies (Gob'nda), &c. , form an important element in the fresh-water fish fauna, 

 but there are no peculiar fresh-water families. 



In the rest of the world, the fresh-water fish fauna consists mainly of families 

 which are confined to fresh water, and which we have every reason to believe 

 have evolved their genera and species in fresh water ; such fishes are of the 

 highest importance of evidence of former land connections or of ancient lines 

 of severance. 



The dominant group of fresh-water fishes is the order Ostariophysi, including 

 the cat-fishes, characins, electric eels, carps, loaches, &c. — not fewer than three 

 thousand species, all true fresh-water fishes except the cat-fishes of the groups 

 Ariime and PtotosidaR, which are (secondarily) marine, and which alone cross 

 Wallace's Line eastwards. 



In Borneo 3 the true fresh-water fishes are mostly Ostariophysi. Of this 

 order thefe are approximately — 



120 species of Cyprinidon 



>.'oiie of these families is represented in Celebes, neither are the Nandida (five 

 species in Borneo) nor the Mastacembelidm (seven species in Borneo). What then 

 are the fresh-water fishes which give Celebes its Indian character? They are, 

 first of all, Labyrinthic fishes, found in the Ethiopian and Oriental Regions, 

 and represented in Borneo by — 



1 species of Lucioccphalida 

 12 ,, .. Anabantidce 

 20 ,, ,, Ophiocephalidce 



Two of the commonest Indian species, Anabas scandens and Ophiocephalvs 

 striatus, not only cross Wallace's Line, but Weber's Line, occurring in Celebes, 

 the islands from Lombok to Timor, Amboina, and Halmahera. It is almost 

 inconceivable that these islands can have been connected with each other and 

 with Asia during the life-time of these two species, yet Professor Max Weber 

 contends that they have not been introduced, as he found them in places where 

 the population was not sufficiently civilised to have introduced them. This 

 contention loses its force when we remember that the Labyrinthic fishes are 

 remarkable for the time they can live out of the water and for their habit of 

 migrating overland from one pond or stream to another, and that these two 

 species are greatly appreciated as food by the Malays and the Chinese. 



Two Indian Symbranchoid eels, Monopterus jaranensis and Symbranchus 



1 Peleeneer, Bull, de V Acad. roy. de Belgique, 1904, p. 1001. 



2 Zool. Ergebn. lieis. Ned. Ind., iii., 1894. 



3 C'f. Popta, Notes Leyden Mus., xxvii., 1906. 



