120 Mr. G. A. K. Marshall on Biaposematisjn, with reference 



slight characteristics which distinguish halicnsis from the 

 typical race rcinwardtii are essentially such as we should 

 expect to find in a group of individuals which have been 

 comparatively recently isolated from the main body of the 

 species ; in other words, baliensis differs from rcinwardtii 

 simply because it has been confined to the island of Bali. 



If we examine such a series of forms as Hwphina phryne, 

 ncrissa, iichcncsa and corva, it seems clear that we are 

 dealing with those progressive modifications which are 

 generally comprised under the name of geographical races ; 

 that is to say, the differences exhibited are accepted as due 

 to the influence of either climatic causes, or isolation, or a 

 combination of both. It is the more probable that such 

 is here the case when we find that the allied Indian 

 H. oiadina ^ presents similar modifications; being repre- 

 sented by an intermediate form, andamana, in the 

 Andamans, and a more heavily-bordered form, fawcetti, 

 in Sumatra. Such progressive widening of the black 

 borders may also be observed in other Pierine genera, 

 such as Delias, Frioncris, Apjnas, etc. ; while from Africa, 

 and doubtless from many other parts of the world, 

 numerous parallel cases could be cited in which no 

 mimetic interpretation could cover the facts. 



Further, it may be mentioned that heavy black borders 

 are a very common feature in the genus Hnphina and 

 exist in a majority of the species occurring in the Malay 

 Archipelago. They attain their highest development in 

 aj/inis (Celebes), in which they occupy nearly half the 

 wing and are far broader than anything to be found 

 throughout the whole genus Ixias. A considerable 

 number of species from these islands could be mentioned 

 in which the borders are markedly better developed than 

 in H. corva. 



I find myself unable therefore to accept the suggestion 

 that the broader black margin of H. corva is due to the 

 direct mimetic influence of /. baliensis ^. It is possible 

 that the Icnas may have mimicked the Hvphina, but in 

 dealing with such black and white Pierines a hasty 

 assumption of mimicry is specially to be deprecated ; and 

 it is w^ell to bear in mind the judicious warning in this 

 connection uttered by Wallace forty years ago : " By far 

 the most general type of colouring in the Pieridse, and 

 which recurs in hundreds of species, is a white ground 

 with a black outer border, always most developed at the 



