ILLUSTRATIVE OF NATURAL SELECTION. 149 



indication of how dimorphism may be produced ; for 

 let the extreme Philippine forms be better suited to 

 their conditions of existence than the intermediate 

 connecting links, and the latter will gradually die 

 out, leaving two distinct forms of the same insect, 

 each adapted to some special conditions. As these 

 conditions are sure to vary in different districts, it 

 will often happen, as in Sumatra and Java, that the 

 one form will predominate in the one island, the 

 other in the adjacent one. In the island of Borneo 

 there seems to be a third form; for P. Melanides 

 (De Haan) evidently belongs to this group, and has 

 all the chief characteristics of P. Theseus, with a 

 modified colouration of the hind wings. I now come 

 to an insect which, if I am correct, offers one of the 

 most interesting cases of variation yet adduced. Pa- 

 pilio Romulus, a butterfly found over a large part of 

 India and Ceylon, and not uncommon in collections, 

 has always been considered a true and independent 

 species, and no suspicions have been expressed regard- 

 ing it. But a male of this form does not, I believe, 

 exist. I have examined the fine series in the British 

 Museum, in the East India Company's Museum, in 

 the Hope Museum at Oxford, in Mr. Hewitson's and 

 several other private collections, and can find nothing 

 but females ; and for this common butterfly no male 

 partner can be found except the equally common P. 

 Pammon, a species already provided with two wives, 

 and yet to whom we shall be forced, I believe, to 

 assign a third. On carefully examining P. Romulus, 



