ORIGIN OF VARIETIES IN BUTTERFLIES. 951 



undergo tlie cycle of their cliangcs luove tlian once n year, and thus oj)j)or- 

 tunities are not only niulti[)lied but varied. 



Ordinary variation, due to unknown or diverse causes, as well as that 

 which springs from latitudinal range and distinct climatic influence, ap- 

 pears in butterflies as iu other creatures. In these cases we suppose 

 advantageous variations to be perpetuated and intensified by the survival 

 of the fittest, through the laws of inheritance. By slow accretions, a 

 species multiplies into varieties, each departing from the other and from 

 the original tyvie, until all become firndy fixed as species, again to 

 undergo division. Now just as tiie climatic influences of latitude ap- 

 pear to be an important factor in the development of new forms, so 

 the difference of the seasons may work similar alterations in double 

 brooded butterflies ; we have merely to suppose the zebra swallow-tail, for 

 example, to hibernate exclusively in the imago state, to fix the variety 

 I. a. ajax as the only form that will survive ; on the other hand, let the 

 insect hibernate as now in the chrysalis and be only single brooded, and 

 this form would become extinct ; suppose again both features to liold with 

 different sets of individuals, gradually comnumicatiug this tendency in 

 greater and greater force to tlieir offspring, and we should behold the 

 spring and summer varieties changed to separate species. This is one ex- 

 ample of a mode in which seasonal dimorphism may become an originator 

 of new forms. It is plain that entirely similar results may follow from 

 unequal letliargy in one brood of caterpillars, such as we find in the 

 fritillaries and the crescent spots. 



Ordinary dimorphism again, or the appearance of different varieties in 

 each brood, running through both sexes, must surely be a precursor of a 

 division in the species; no doubt the change is gradual, so that cen- 

 turies of direct experimentation would throw no light upon the change ; 

 but we have only to suppose each form breeding true to itself and the 

 eeparatitm w-ill be accomplished. In the case of the violet-tip, we have 

 two strikingly different forms, which may be distinguished, not only by 

 their coloring and markings, but even by the form of the wings and the 

 sculpture of the hai'd pai'ts of the abdomen ; in fact, we hav-e two forms, 

 permanently distinct from each other, to which we cannot apply the name 

 of species simply because we know they have the same immediate parent- 

 age ; we can hardly doubt that the separation of this species is nearly 

 accomplished. 



The same may be aflSrmed of antigeny ; we find melanic antigeny both 

 partial and complete ; in its partial condition in our spring azure it grows 

 more and more nearly perfect as we pass southward : we see its form fixed 

 in some species, and in others the melanic feature has been superinduced 

 upon the opposite sex. The same is true of structural antigeny. Some 

 species, which we can hardly doubt have had a common ancestor, scarcely 



