106 Dr. Arthur G. Butler on 



satisfied that in an unusually dry country the so-called 

 " wet-season form " is sometimes identical with the so- 

 called " dry-season form " of a moister country. Indeed, 

 the conditions, as far as the absence of one modification 

 and presence of a new one are concerned, are practically 

 very similar to those which obtain in the N. American 

 Lyca3nid, Cyaniris pseudargiolus, beautifully worked 

 out by Mr. W. H. Edwards (Butt., N. Am., ii., pp. 315 

 et seq.). 



Mr. Marshall is quite correct in saying that it is 

 extremely difficult to define " the specific differences of 

 butterflies merely from a series of museum specimens 

 when not backed up by a practical knowledge of 

 the habits and range of the species involved." But, 

 on the other hand, it is impossible for a worker in any 

 one part of so vast a country as Africa arbitrarily to 

 settle, to the satisfaction of everyone, the extent of varia- 

 tion of any one widely distributed species under seasonal 

 and climatic conditions. That Mr. Marshall has amply 

 proved this I can readily show ; and J do so, not with 

 any desira to detract from the value of his observations, 

 as applied to South African butterflies when in southern 

 S. Africa, but to prove that the conditions differ in 

 the same species when found only so far northward as 

 Nyasaland. 



In my notes on {( Seasonal Dimorphism in Acrsea " I 



took what I regarded as the extreme dry and wet 



developments of the species, not of that phase of the 



species represented either in the Cape Colony or in 



Mashunaland ; whether the intermediate forms occur as 



the dry- season form in one part of Africa, the wet-season 



form in another, as the sole representatives of the species 



in a third, or not at all in a fourth, is perfectly immaterial. 



As developments of the species, intimately connected 



with and incapable of separation from it, they must be 



taken into account ; but I frankly admit that it is 



extremely probable, as already granted, that I ought to 



have called the dry-season form " wet/' and the wet form 



" dry/' The fact, therefore, that one of ray seasonal 



forms does not occur in conjunction with the extreme 



southern form throughout its range, or that it occurs 



apart from the latter at any part of its range, is not 



enough to show that it is not a seasonal phase of a species 



in some part of Africa. 



