96 LEPIDOPTERA. 



locally as the perfect insects ; and then we have many 

 instances where the larvae of species, which are really quite 

 distinct, are so like one another that we do not well know 

 how to distinguish them. 



Since in this genus we have especially to deal with the 

 foi'mation of local varieties (races), I cannot forbear from 

 entering a little upon that subject. The theory of varieties 

 is grounded on facts in various branches of natural history. 

 This is perhaps most strikingly the case in plants, where we 

 can often prove from the seeds taken directly from them that 

 forms of very different appearance truly belong to one species, 

 and when also we generally know, more or less correctly, 

 what are the modifying agents. This is, however, much 

 less easily the case in Zoology, and probably specially so in 

 Entomology; yetsitleast in Lepidopt era we have a number 

 of indubitable cases of varieties. In order to prevent all 

 misconception, I must here repeat what I have already said 

 in the preface to our new Catalogue of Lepidoptera, that I 

 emphatically distinguish between varieties and aberrations, 

 and by the latter I understand only accidental modifications 

 of a species, which are quite independent of time and place. 

 The varieties I divide into those dependent on time and those 

 dependent on locality. It is known that very many Lepi- 

 doptera, though perhaps only in some places, have two or 

 more generations in the course of the year. So, for example, 

 PoJyommatus Phlceas, in the north of Europe (in Finmark), 

 is only single-brooded, whereas in central and southern 

 Europe this species is double-brooded. It is also known 

 that these different broods have often such an influence on 

 the species, that each of the broods was long reputed to be a 

 distinct species. The best known example of this is Vanessa 

 Prorsa, L.,and Levana, L., whereby breeding from the egg 

 their specific identity was proved to demonstration. And I 



