MOEPHOLOGT OF THE LEPIDOPTEEOUS PUPA. 191 



larval tissues in a state of histolysis, from -whicli the imago will be rapidly built up 

 at a later period. 



Sir John Lubbock * has suggested that the explanation of the quiescent pupal stage 

 is to be found in the difficulty with which a gradual transition could be effected from 

 the biting mouth-parts of the larva to the sucking mouth-parts of the imago. The inter- 

 mediate form of mouth would be unfitted for either biting or sucking successfully; 

 hence the necessity for a quiescent stage in which no food is taken, and during which 

 the change can be accomplished. 



The principle which underlies Sir John Lubbock's suggestion has probably been of 

 very great importance for other structures as w^ell as the mouth-parts, viz. the fact that 

 specialization to the conditions of life in the final sexual phase of existence is thus 

 readily attained without interfering with the great specialization of other antecedent 

 stages. It would be of the greatest importance for the imago to be able to modifv the 

 methods by whicli it is adapted to its environment, without the cumbrous necessity for 

 such modifications to bo gradually introduced through a number of previous stao-es. 

 This is the explanation of the quiescent pupal period adopted by the late ~E. M. Balfour f. 

 A general support to this argument is to be found in the fact that the difi'erences between 

 larva and imago are far less in the Orthoptera, with their gradually progressive meta- 

 morphosis. The difficulty of specialization to different conditions in different stages has 

 here been met by uniformity in the stages, so that one form of specialization is, with 

 slight differences, available throughout. 



I will briefly recapitulate the history of metamorphosis, which is, I believe, supported 

 by the facts to be described in this and succeeding papers. One of the terminal stages 

 of a gradually progressive metamorphosis, sufficiently advanced to possess well-formed 

 external reproductive organs, became quiescent ; this stage is largely preserved in the 

 external morphology of the pupa. Histolysis and re-development of the final form 

 then occurred, gradually displacing the stages immediately antecedent to the quiescent 

 phase, which previously had been necessary in order to lead vip to the latter and the 

 final form to which it gave rise. Becoming more and more complete, the changes 

 beneath the surface of the i^upa gradually displaced the earlier stages until only the 

 first, the larval stage, remained. The morphological intervals between pupa and larva 

 and between pupa and imago have been subsequently widened by specialization to 

 the conditions peculiar to eaclx stage. The comparatively slight differences between 

 pupa and imago are entirely to be explained in this way ; for very few, if any, stages have 

 been omitted between them. 



These conclusions apply to the Lcpidoptera, and probably with slight modification to 

 other orders with a similar form of metamorphosis. 



Part I. — The persistent Traces of Larval Structures upon the Pupa. 



Before considering the number of abdominal segments and their relation to those of 

 the larva, it is necessary to describe certain characters which will form very convenient 



* ' Origin and Metamorphoses of Insects.' t ' Comparative Embryology.' 



