90 Bulletin 64. 



A. study of the adult moths shows little change except in the 

 antennae, wings, and genitalia, but in neither antennae nor wings 

 are there variations enough to lead us to regard them as the prime 

 characters which separated any species. In the immature stages 

 there is a great similarity among all the species studied. This 

 similarity among the species in both adult and immature stages 

 may be explained by the great similarity of habit. So far as is 

 known the species of Cra»ibus are grass insects. Throughout the 

 growing season the grass offers practically the same conditions ; 

 hence great variation in the organs concerned in the function of 

 nutrition could hardly be expected. The species are so small and 

 require such a relatively small amount of food, that many indi- 

 viduals can exist upon a small area. The grass is so open and the 

 moths fly so readily that there is little need of highly specialized 

 organs to facilitate the finding of a mate ; consequently in the 

 secondary sexual organs we find a rather uniform degree of devel- 

 opment. 



In the male genitala there are great variations. These varia- 

 tions may be explained by the almost uniform abundance of food 

 throughout the growing season. If grass is kept cut or grazed 

 rather close to the ground, as it frequently is in nature, it grows 

 throughout the season. Under these conditions the insect that 

 fed continously throughout the growing season, and was not 

 more destructive at one period than another, would succeed best 

 in the struggle for existence. Like many larvae, Crajnhis larvae 

 are more destructive just as they are completing their growth. A 

 genus of insects might become adapted to these conditions by 

 having several generations in a season or by breaking up into a 

 number of species, each having a different breeding season. In 

 Crambus there are a number of species in a locality breeding once 

 •a year, but at different times. Among the primitive Crambids 

 the abundance of food at other than the usual period of larval 

 existence would tend to preserve any variations in the time of 

 breeding. Those individuals exhibiting the greatest variations 

 in the time of breeding would naturally pair, and should any 

 structural variations arise whereby pairing with individuals con- 

 forming more closely with the usual breeding time might be ren- 

 dered difficult, such variations would be preserved by natural 

 selection. A slight variation in the male genitalia might render 



