MELANISM AND MELANOCHROISM. 299 



uniformly cold in the north and warm in the south, these 

 seasonal forms would be transformed into geographical races, 

 and ultimately species. Such may have been the origin of 

 G. rhainni d,x\(\ cleopalm'' {Entomologist, vol. xxii., pp. 27-29). 



I must say that I do not altogether follow Mr. Cockerell's 

 reasoning here as to " growth " in the pupal stage. The 

 foregoing paragraph certainly leaves me with the impression 

 that Mr. Cockerell believes some change in the size of the 

 imago may take place by growth, in the pupal stage. I 

 believe myself that the size of the imago is entirely determined 

 in the larval stage, and fixed as soon as the larva becomes a 

 pupa ; but that different degrees and rates of development may 

 affect not only the condition of the membranous structure of 

 the wing, but the scale structure as well, is very probable, and 

 whilst not at all prepared to criticise Mr. Cockerell's state- 

 ments as regards development and its effects, I do most cer- 

 tainly believe "growth," as determining size, impossible in that 

 stage. I have before hinted that, possibly, delay in the pupal 

 stage may to a small extent help to bring about such changes 

 as occur in seasonally dimorphic species (although I believe all 

 the greater changes result from delay, etc. in the larval stage) ; 

 and since cold produces delay both in the larval and pupal 

 stages, cold may thus indirectly influence the colour of insects 

 to a very slight degree. But I cannot leave out of account the 

 fact, that in nature, cold does not have the least tendency to 

 produce a dark coloration, and that the arrest accompanying 

 extreme cold tends to produce, as a rule, pallid ill-marked 

 specimens, with a tendency to become generally unicolorous, 

 apparently the result oftentimes of ill-developed pigment in 

 the scales. 



With regard to seasonal forms in Britain, we may consider a 

 few cases. Pieris rapce. The early brood (May and June) speci- 

 mens are much whiter with us than those of the summer brood. 

 I think this affords one of the most striking proofs of the absence 

 of the effect of cold on the pupal stage. Pieris rap(2 lays its eggs 

 in August and September, pupates in September and October 

 (frequently in larval stage under four weeks), remains as pupa 

 exposed to cold all the winter, and emerges in May as a pale 

 form, frequently with the dark spots obsolete and therefore 

 perfectly white. Eggs of the summer brood are laid in May, 

 larvae feed through [une and July, pupate July and August, 

 • emerge August and September, the imagines being compara- 

 tively dark, having dark tips to wings, and distinct black 



