Geographical Distribution of European RJiopalocera. 491 



Genus Cyclopides. 



The three European species extend eastwards as far as Amoor- 

 land. 



I am sorry I cannot give the distribution of the species north 

 and south with more accuracy, owing to my having no catalogues 

 of the butterflies of North Africa and Southern Asia. But we 

 obtain from such species as I have noticed some important results 

 respecting Corsica and America. Insects have travelled between 

 Asia-Europe and America by two routes, one in the direction of 

 Iceland and Greenland, and one through the Aleutian islands. 

 The species that are common to Asia-Europe and America have 

 almost invariably taken the latter route, thus proving that Europe 

 and North America have been geographically separated for such a 

 length of time as to allow for all the species that once travelled 

 between them to vary to such an extent as to .be reputed distinct 

 in the two countries, for only Pyrarneis Atalanta and a few 

 other powerful species of that kind have remained unchanged out 

 of the numbers that must have travelled that route. If geo- 

 logists can tell us how- long Corsica has been separated from 

 the main land, we shall have some most valuable data by which 

 to calculate the length of time that the formation of a species 

 requires, for a large number of Corsican insects are already good 

 species, and many more have become sub-species, which is the 

 last step towards the formation of a new species, especially when 

 the variation is correlated in the larva, as it is in Vanessa Ichnusa. 

 In conclusion, I do not think any one present, however much he 

 may differ from the opinions expressed in this paper, will regret 

 my bringing prominently forward Mr. Darwin's theory of the 

 Origin of Species, for it is the great scientific question of the day, 

 and ought to be freely discussed here as well as in other societies ; 

 and my own opinion is, that, in a great measure, by the study of 

 geographical distribution in its widest sense, the theory must stand 

 or fall. 



