FORM, GROWTH, AND CHANGE 63 



pupa by the tail is alone required. Spinning concluded, the 

 body of the caterpillar becomes much shortened and the 

 larval cuticle, splitting lengthwise in the mid-dorsal line, is 

 slowly worked off the body from before backwards. Thus the 

 form of the pupa (Figs. 29 e, 30 B, 33) is revealed, and examina- 

 tion shows that it prefigures all the characteristic structures 

 of the butterfly. Wing-rudiments (Fig. 33 w, 35) of the same 

 subtriangular shape as the butterfly's forewings, although 

 smaller, lie on either side of the body ; on each side of the head 

 is the area of the compound eye ; the mandibles, so prominent 

 and important in the caterpillar, have disappeared, but the 

 long, jointed maxillary galeae which make up the sucking 

 trunk of the imago, lie side by side stretching backwards from 

 the head beneath the ventral region of the thorax between the 

 feelers (Fig. 33 /) and the fully-segmented legs which are 

 packed between the costal edges of the wing-rudiments. The 

 abdomen covered like the rest of the pupa by a cuticle 

 which soon becomes firm and largely rigid is much shorter 

 than the same region in the caterpillar ; its segmentation is 

 well-marked, and the presence of flexible cuticle at three or 

 more of the intersegmental junctions allows a restricted degree 

 of movement. The tail-end is pointed, the spinose extremity 

 (or cremaster) (Fig. 33, cr) serving to anchor the pupa to its 

 cocoon or its suspensory pad. 



Consideration of the apparently sudden production of these 

 organs at the pupal stage suggests some important questions. 

 Have they indeed been then formed for the first time, and if 

 not, why is no trace of them to be seen through all the changes 

 of the caterpillar-life ? The answer leads us to appreciate the 

 inwardness of the butterfly's type of life-history. Thinking 

 especially of the wing-rudiments, so conspicuous externally 

 on the nymphal grasshopper or dragon-fly, we dissect a half- 

 grown caterpillar and find within the second and third thoracic 

 segments little flattened pads (Fig. 34 ef), into which air-tubes 

 run, lying in pouches connected with the body-wall. 1 These 

 are the wing-rudiments, yet though dissection is needed to 

 demonstrate their existence, they are not, properly speaking, 

 internal organs. Study of a series of sections shows that they 



1 J. Gonin : " Recherches sur la Metamorphose des Lepidopteres ". 

 Bull. Soc. Valid. Sci. Nat., XXXI. 1894. 



