I50 IXSKCTS Axn ]if.ri:ditv 



colouring.' Both larva and pupa are freely exposed to 

 view dirough chinks in the scant)- network and between 

 the imperfectly united leaves. Uropteryx sanibiiiaria (the 

 * Swallow-tailed Moth') constructs a slight hanoino- cocoon, 

 aftbrdino- very little concealment. The enclosed pupa 

 bears a marked cr\ptic appearance, while the onl)- experi- 

 ment which has been made indicates the possibility of a 

 well-developed power of individual colour adjustment.-' 

 Both these species, and especially the last, have long since 

 reached the staije at which the reduction of the cocoon 

 became advantageous. In the genus ZonosoDia {^Jipliyra) 

 we inde|)endentl)' arrive at the same climax of reduction 

 attained in the Picriiiac, «S:c.. the cocoon being repre- 

 sented b) a supporting loop and the means of fixation of 

 the posterior extremit}.'' No search has been made, so 

 far as I am aware, for a special strengthening of the 

 cuticle upon which the loop presses, but it is probable 

 that nothing of the kind is required by these small light 

 pupae. The exposed Hphyrid chrysalis is fully as cr\ptic 

 as that of the average butterfly, l)ut it lacks the power of 

 colour adjustment. When the Eph) rid larva is dimorphic, 

 green or l)rown, the colour ol the pu|)a corresponds 

 to that of the larva from which it develojjed."* Such 

 correspondence has not been observed in any other 

 Lejjidopterous insect. 



If we take into accotmt the fact that Zonosonia {^lipliyra) 

 is a characteristic Geometrid oenus, althouHi its method of 

 pupal susi-)ension is unique in a family whose species niake 

 cocoons or bur\-. we may feel confident that it has been 

 descended from cocoon-making ancestors, and that Etigo- 

 nia qiierciiiaria and i roptcryx savibucaria give us a clear 

 idea of the steps by which the reduction was effected. 



' Trans. Eni. See. Lend., 1885, p. -519. 



' See Colours of Animals, Loiulon, 1890, pp. iii, 112. Only one 

 examj)le was placed on while paper before pupation. Although ihe 

 resulting chrysalis was very pale and strikingly difTerent iVoin the ordinary 

 appearance, the evidence is quite insufficient, and it is much to be hoped 

 that the exjieriment will be repeated upon a large scale. 



' Trans. Ent. Soc. J.ond.. 1884. p. 57. 



^ Trans. Ent. Soc. Lond.. 1884, p. 51 : Phil. Trans. Roy. Soc.^ vol. 178 



(1887), B., pp. 437, 438- 



