The Bionomics of South African Insects. 501 



closely related to the Nymphalinm that it is difficult to 

 draw a line between them. The argument is not of much 

 weight, because the intensely procryptic habits and colours 

 of many Nymphaline genera have certainly been brought 

 about by selection due to the great keenness and suc- 

 cess of insect-eating animals in their pursuit, I have 

 however suggested and brought evidence in support of the 

 view that some of the procryptic Nymphaline species are 

 to a certain extent unpalatable (see p. 442). 



2. Mimicry in the Nyniplialinie does not appear in 

 isolated forms but in all or nearly all the species of a 

 genus. Such mimetic genera are usually very large, 

 dominant, and wide-spread. The species themselves are 

 also often wide-spreading, and may have an enormous range 

 far exceeding that of the model {Tlijiiolhnnas misippus). 

 Allowing for the fact that the mimetic species resemble 

 the commonest types in the world, and so are liable to 

 escape notice, it is probable that they are rich in indi- 

 viduals. In many instances we know that this is so. 

 The more we investigate it the more does Rhopaloceran 

 mimicry seem to be associated with dominant genera and 

 species, rather than the feeble and hard-pressed forms 

 which H. W. Bates presupposed in his well-known tlieory. 



3. The dominant tendency towards mimetic resem- 

 blance in any genus cannot be explained by hereditary 

 transmission of the mimetic form of a single parent species, 

 or from the tendency of closely-related species to vary along 

 nearly the same lines, because the species of a mimetic 

 genus, as a matter of fact, mimic in many different direc- 

 tions. Thus Pseudacrrca resembles Acrx,a, Planenia, 

 Amauris, and Limnas cJirysipipus ; while Hypolimna^, 

 including Euralia, is even more protean. 



4. The non-mimetic species of a mimetic genus are 

 often markedly conspicuous, exhibiting what has all the 

 appearance of an aposematic pattern peculiar to them- 

 selves {Rypolimnas, Psendacvica). This is also frequently 

 true of the non-mimetic males of a species with mimetic 

 females {Hypolimnas). Such aposematic patterns are 

 especially displayed on the under-side, where procryptic 

 colours are developed in other butterflies. 



5. The converse of the last argument is also true, viz. 

 some of the species in a genus, which is as a whole 

 markedly conspicuous and itself mimicked, are often 

 mimetic of quite other groups. Many instances oi Neptis 



