142 THE entomologist's record. 



sents vermiform stages that the Orthopteron passed within the egg. 

 The other is that it is descended from, and represents, the active pre- 

 imaginal states of the Orthopteron. That there is no retrogression in 

 the sense that the other theory assumed, but merely modifications of 

 various parts to meet various needs, modifications amounting in some 

 cases to temporary suppression of parts and their reduction to 

 imaginal discs, i.e., collections of embryonic cells, undergoing no 

 development until a very late stage in the life-history — this being, 

 nevertheless, no retrogression, but rc:illy an advance. That this latter 

 theory is correct seems proved for us by the crucial examples aftbrded 

 us in the parasitic beetles, Meloc, Mijlahris, Metoccit^, and others, as 

 well as other forms such as Sti/Iop:^, which is perhaps truly a beetle. 

 These hatch with good articulated legs, which they afterwards entirely 

 lose and regain as imagines. In these, therefore, the vermiform stage 

 is posterior to the articulate ones, and cannot, therefore, be descended 

 from a stage anterior to that possessing limbs. 



When larval limbs are entirely absent the imaginal limbs exist in 

 the larva as imaginal discs. How do they exist in the typical lepi- 

 dopterous larva with S-jointed corneous legs '? 



It is necessary to clear the ground a little, and ask ourselves what 

 we definitely mean by this question, and what alternative answers may 

 be or have been given. On the one hand Ave observe that in an 

 Orthopteron at each moult the new leg is formed within the old one, 

 and is withdrawn from it at the moult, or, more accurately, the new 

 leg is the old one enlarged and grown, and freed from the external 

 chitinous sheath, which had grown too small for it. Is the lepi- 

 dopterous leg in this case ? Is the imaginal leg really the larval leg 

 grown and developed ? 



On the other hand, in the apod larva the imaginal leg is an 

 imaginal disc. Is this also the case in the lepidopterous larva, the 

 larval leg being got rid of as a larval appendage that has completed its 

 life cycle and its usefulness, the new or imaginal leg entirely develop- 

 ing from an imaginal disc within the body of the caterpillar ? 



I believe the former is the true answer, the larval leg is the 

 imaginal leg also ; but this answer must be modified to this extent, 

 that the imaginal leg is so far different from the larval one, that it is 

 to a great extent a new development, arising, therefore, to that extent, 

 from imaginal discs, i.e., reserved embryonal cells, but with this im- 

 portant difference from the second ansAver I have suggested, that they 

 are situated each within its OAvn segment of the larval leg, or even 

 more definitely, each in association Avith its OAvn portion of each 

 segment. The imaginal tarsus arises from the actual larval structures 

 of the third joint of the larval leg and from embryonal cells situated 

 amongst them, the imaginal tibia in the same Avay from the larval 

 tibia, and so on, the trochanter and coxa being almost entirely in an 

 embryonal state at the base of the leg. This ansAA^er, which I support, 

 is that that has been long held by the feAV entomologists Avho have 

 paid any attention to the matter, and Avas believed to be supported by 

 certain experiments of Reaumur's* and NeAA'portf, as to the effect on 

 the imaginal structures of removing the larval legs. 



* Reaumur, Memoiyes des Insectes, i., p. 365 (1734). 

 f Newport, Phil. Trans., 1844. 



