308 Mr. A. R. Wallace on 



In my paper on the eastern PapUion'td^ (Trans. Linn. Soc. 

 XXV. 1) I called attention to the curious form of wing that charac- 

 terizes most of the species of the island of Celebes. In the 

 Fieridce it is almost equally prominent, as may be seen by com- 

 paring the following species, nine in number, with the most nearly 

 allied forms from the adjacent districts : — 



Pieris eperia, as compared with P. coronis. 



Tlujca zebuda, „ „ T. Dcscombcsi. 



Tityca Rosenberg'ii, „ ,, T. hijpnrete. 



Tachyris Ilombronii, „ „ T. lynclda. 



Tachyris lycaste, „ „ T. lyncida. 



Tachyris zarinda, „ „ T. nero. 



Tachyris ithome, „ „ T. nephele. 



Eronia tritcea, „ „ any other species. 



Iphias glaucippe, „ „ I. glnucippe {Java \ar.). 



In all these cases, comprising a large proportion of the Pieridce 

 peculiar to Celebes, the species of that island have the anterior 

 wings either more elongated or more acute, or the costa much 

 more abruptly curved near the base, than in the allied species 

 from any of the adjacent islands. Exactly the same peculiarity 

 was found in almost every species of Papilio, and is to be noticed 

 also in many of the NymphalidcB. I am not aware that any 

 similar case has yet been adduced of a peculiarity of form runnin;^' 

 through several distinct families, and characterizing a single island 

 only, or any equally limited district. The nearest approach to it 

 is the apterous condition of many Coleoptera of usually winged 

 genera in Madeira, conjoined with the rather fuller development 

 of the wings in some other species, whose habits render flight 

 essential to them. We are all acquainted with Mr. Darwin's 

 beautiful explanation of these facts, on the principle of the 

 preservation of useful variations, and we cannot doubt that the 

 same principle has acted in the case of the Celebes butterflies, 

 although from our ignorance of all the physical and biological 

 peculiarities of that island we are not able to trace the process 

 with so much certainty as in the other case. The whole fauna of 

 Celebes is isolated and fragmentary, and indicates a very ancient 

 origin and much extinction of species; and it may well be that 

 the conditions which once rendered this peculiarity an advan- 

 tageous one no longer exist. We must ever remember that the 

 present condition of a fauna is but the resultant of an immense 



