The Bionomics of South African Insects. 459 



exercise a wonderfully calming and steadying influence. 

 The structures which are adopted as the conventional 

 criteria of specific distiuction are of course modified by 

 natural selection and brought into adjustment with new 

 conditions of the struggle for existence as one species is 

 gradually changed into another; but tliey are also capable 

 of modification in one and the same species as it passes 

 through various conditions during its life-history and in 

 sexual and other dimorphism. The species frequently 

 requires that the female sex should be more protected than 

 the male, and hence we often witness a more perfectly 

 cryptic appearance and habits in the female, and mimicry 

 in the female alone. In many kinds of di-, tri- and 

 polymorphism we see a species more perfectly protected 

 at one and the same time by extending the area over 

 which it must be sought by its enemies — in cryptic 

 resemblance, earth and bark as well as leaves and shoots — 

 in mimetic resemblance, Danaine or other distasteful 

 models not of one species alone but two or more. In the 

 di-, tri- or polymorpiiism of the social Hymenoptera and 

 Neuroptera we see the specialization of the individual for 

 the good of the community. In the extreme cases of 

 seasonal dimorphism, exhibited by the genus Precis, there 

 is a far less common modification of a species into two 

 series of generations respectively adjusted to the con- 

 cUtions obtaining at two seasons of the year. But less 

 marked cases of the same kind are probably not imcommon. 

 There is however nothing revolutionary or subversive in 

 any of these interesting facts, ' The conventional marks of 

 specific distinction remain just as they were, convenient 

 indications to the systematist, enabling him provisionally to 

 separate groups of individuals into the assemblages we call 

 species. When his work is done carefully subsequent 

 breeding experiments will, we may be sure, confirm his 

 conclusions in the majority of cases. But here and there 

 startling exceptions will be found when it is to the 

 a'lvantage of a species to appear in two or more very 

 different forms. In such cases the reason for the differ- 

 ence can generally be satisfJictorily explained on the 

 principles of natural selection ; and when such an explan- 

 ation is possible or even probable it is quite unnecessary 

 to assume that the exceptions possess a numerical import- 

 ance sufficient to shake the foundations of systematics. 

 Certain species are cryptic while others are aposematic 



