158 MR. C. N. BARKER 



as that it is the result of miniicrj by natural selection? 

 Some very instructive results might be obtained if sys- 

 tematic experiments were made, by transmitting the ova 

 of species from one district and breeding from them in 

 contiguous districts, under as natural conditions as pos- 

 sible, through many generations. 



In October, 191B, I received from Mr. H. Barber a small 

 collection of butterflies all taken during March of that 

 year at Kampala in Uganda. Included among them are 

 specimens of Ergolis inotria, Cr. (Nymphalinre) and 

 Ypthima argentata, Bartol (Satyrinse). These two 

 species of distinct sub-families are covered with a surface 

 scaling of a whitish silvery colour which covers and ob- 

 scures the sombre browns peculiar to their congeners. I 

 think this can be safely claimed as a case of similar en- 

 vironmental results on two widely separated species 

 brought about in the way I have sketched above. 



The butterflies of the Celebes islands have acquired a 

 similarity in the elongation of the forewings. Can this 

 peculiarity be claimed as the result of mimicry? If so, 

 Avhat benefit do they derive from this abnormality which 

 affects the Avhole community? Philocnema latipes, de 

 Geer, and many other Callichromides vary from green to 

 purple in the colour of their elytra according to the lo- 

 calities which they frequent; in Natal they are green on 

 the coast belt and become purple in the upland districts. 



Among the butterflies of South Africa there are several 

 groups which come under the Mullerian formula of 

 Mimetic Associations, which have been designated by 

 Prof. Poulton as '/ Synaposematic " groups. The prin- 

 cipal models for these groups are included among the 

 Danaina3 and the Acrfeinne. Other large mimetic associa- 

 tions have been proposed, made up of insects of many 

 denominations that have a common system) of coloration, 

 and in these groups are included insects of all shapes and 

 sizes. Chief among these can be mentioned the Lvcoid 

 and Multilloid groups. As a colour group the Pieridae 

 are by far the most conspicuous and numerous, yet, some- 



