Origin of the Fauna of Celebes. 131 



Here we are dealing in part with large mammals. This 

 difference between North and South is all the more marked 

 that the number of mammals of Celebes is in any case not 

 large. Still more marked are the differences between the 

 land-snails of North and South Celebes. According to the 

 statements of Ed. v. Martens * only sixty-four species had 

 been recorded up to 1891, and of these only two are common 

 to North and South, "namely Trochomorpha planorbis, which 

 is distributed over the other large islands, and Amphidromus 

 perversus, which occurs in Borneo ; as a third there may 

 perhaps be reckoned Nanina limbifera, which is, of course, 

 characteristic of Celebes. . . ." Twenty-seven species are so 

 far exclusively peculiar to the northern peninsula and thirty- 

 one to the south-west. Three species are recorded only from 

 the south-east peninsula and two species from both southern 

 peninsulas. This interesting and remarkable difference in 

 the fauna can be explained only by a former separation of the 

 two regions. The junction of these regions since the Neogen 

 period could not have sufficed to counteract the difference 

 as regards the sluggish land-snails, which are not easily 

 transported, and this difference is still evident among active 

 mammals. Even among birds it is not yet eliminated. But 

 it can hardly hold good for the freshwater fishes, since these 

 are all of marine origin, and being equally distributed by the 

 sea can penetrate into the most different river-systems. 



If now, both from zoological and geological standpoints, 

 important reasons exist for the acceptance of the theory that 

 Celebes consisted formerly of several unconnected islands • if, 

 further, geology makes it probable that this was the case 

 during the second half of the Tertiary period ; then it follows 

 that the present rivers are of recent date, and the small size 

 of the islands previously prohibited the formation of river- 

 systems of any considerable size. Thus we see that the 

 group which we now call Celebes was, in the second half of 

 the Tertiary period, altogether unfitted for the production of 

 a fauna of freshwater fishes. This would seem to me to 

 explain the peculiarities of the fish-fauna of Celebes more 

 satisfactorily than the theory that Celebes was separated from 

 the Indian continent before it could have been peopled with 

 Cyprinidae and Siluridas, for the simple reason that these 

 genera had not yet appeared on the earth. This, however, 

 seems to be Giinther's view, if I am not mistaken. As we 

 see, in his statement quoted above at length, he ends with the 



* E. von Martens, " Landschnecken des Indischen Archipels," in 

 Max Weber, Zoolog. Ergebnisse &c. Bd. ii. 1892, p. 259. An error bas 

 slipped in among tbe numbers of the species in North and South Celebes. 



