18 NATURAL SCIENCE [January 



chrysalis to pass from the type of the Succincti to that of the Sus- 

 pensi. Just as Mr Scudder finds in Hespcria, making a ' shroud ' 

 within its cocoon, the beginning of the Succincti, so in Thais, with 

 its girdle slipped up to the ' nosehorn,' we may hud prefigured the 

 beginning of the Suspensi. In each case we are furnished with a 

 hint and no more of the possible modus operandi by which the 

 cocoon and the girdle have become discarded. And if in the next 

 more specialised butterfly than Tliais we found a Suspensid, we 

 might have Mr Scudder's ideal sequence actually realised. As a 

 matter of fact Farnassius, which is this butterfly, walks back into 

 the cocoon, and what becomes of the slipping girdle of Thais we, 

 that is, I, cannot say. Perhaps these or similar considerations might 

 have occurred to Mr Scudder ; but, when one has so beautiful a 

 theory in hand as is supplied by the ' shrouds ' of the Involuti, one 

 may be forgiven for not looking deeper into the matter. Yet it is 

 one deserving of a little thought. 



Long ago I called attention to the separate nature of the conditions 

 of life attending the larva, pupa and imago, and drew the corollary 

 that these different stages must be studied by themselves, and 

 judged by their own standards. The caterpillar and moth of 

 A2KUeLa go their separate ways, liaving different cares to meet, and 

 what in each is a relatively primary character must be distinguished 

 from what is a relatively secondary. In one sense the larva dies 

 when it changes to a chrysalis, and the chrysalis perishes when it 

 discloses the butterfly. Both throw off their former mode of life 

 and its dangers with the cast skin.^ 



It is probably, then, a fallacy to believe, with Mr Scudder, that 

 there is ' a regular progression from the cocoon of the moths to the 

 almost total absence of the use of silk,' and that the facts imply such 

 a progression. Instances are not rare where the generalised forms 

 spin little or no silk, and the specialised forms, on the same line, 

 make large and complex cocoons. This envelope to the pupa is so 

 clearly an adaptive character that in one, single, upon other charac- 

 ters quite homogeneous group, like the ' Emperor Moths,' it runs 

 through the entire scale, from utter absence to a specialisation 

 seldom attained throughout the Lepidoptera, the hanging cocoons of 

 Callosamia and Attacus. Mr Scudder himself speaks of cases of ex- 

 ception to his theory, ' where the necessities appear overwhelmingly 

 great.' So that a ' biological necessity ' exists, at least sometimes. 

 But the necessities are always great, being ultimately matters of life 

 or death. In the ' Emperor Moths ' it is the more generalised forms 

 that spin no or little silk, and the more specialised that secrete 



1 In my original communication, Mitthcilungen aus d. Roemer Museum, No. 8, p. 

 15, I have shown that the pupal specialisations given by Dr Chapman for the Pierinae 

 do not run parallel with tliose offered by the neuration of the imago. 



