376 THE BIOLOGY OF INSECTS 



correlated with characters that are adaptive. Among creatures 

 such as insects in which reproduction is very rapid and the 

 number of individuals is often enormous, natural selection 

 must be an especially potent agency. For the many 

 characters as to whose utility there is no evidence, we may 

 find some explanation in the facts of mutation and Mendelian 

 inheritance. Unfortunately a tendency has shown itself 

 during recent years for investigators of the different possible 

 factors of evolution to minimise or deny the importance of 

 those factors in which they are not themselves actively 

 interested. " Darwinism " and " MendeUsm " (the termin- 

 ology is suggestive of partisan or sectarian controversy) are 

 in fact complementary rather than rival theories of evolution, 

 for the former helps us to understand, in part, adaptive and 

 the latter non-adaptive characters. Much discussion has 

 taken place as to whether large or slight variations are to be 

 considered the more important as '' raw-material " for the 

 action of natural selection, and many advocates of the 

 Darwinian theory seek to emphasise the importance of 

 slight, " almost imperceptible " distinctions ; this was the 

 later view of Darwin himself, although at one time he 

 thought that what would now be called mutations might be 

 of great value in evolution. A variation whether extensive 

 or slight will, if favourable, be selected for survival, and if 

 heritable be transmitted to future generations. It has 

 already been pointed out (p. 355) that definite segregation 

 in inheritance rather than extent is to be taken as the mark 

 of a discontinuous variation or mutation, and Hale 

 Carpenter has shown (19 14) that small variations in wing- 

 pattern of the mimetic butterfly Papilio dardanus are 

 definitely inherited. 



The limits of all classificatory groups — species, genera, 

 families, and the rest — vary greatly in clearness of definition ; 

 where the distinctions are most clearly marked it is likely 

 that many intermediate forms have died out. Yet we have 

 seen that such a striking character as winglessness commonly 

 shows itself in a single generation. It is interesting to find 

 that sometimes a character that may be regarded as generic 



