Flight in the Homoniorphic Insecta. 381 



ance at one of the earlier changes of skin as slight prolonga- 

 tions of the posterior angles of the dorsal arcs of the two 

 hinderniost divisions of the thorax, the mcsothorax and the 

 metatliorax. These jn'olongations are so many du])licature3 

 or fiattened evolutions of the integument — the chitinous mem- 

 brane that covers them above and below and on the edges 

 being in direct continuity with that which covers the insect's 

 body, being, in fact, part of it, and the intermediate cellular 

 layer which produces this chitinous membrane being similarly 

 continuous with that which underlies the skin of the rest of 

 the insect's body. They increase in size slightly at each 

 successive moult, soon acquiring a definite triangular form 

 and the principal nervure dividing the wing into its two 

 principal arcffi ; but, relatively to the future wings, they are 

 small and insignificant even at the last moult, at which the 

 organs of flight are suddenly developed to their fullest extent. 

 If a wing-rudiment be examined just prior to a change of 

 skin, it is found that its external chitinous covering has sepa- 

 rated off so as to be easily detachable from a new wing-nidi- 

 ment that has formed beneath it, and that this new wing- 

 rudiment lies quite flat within its sheath (as the portion of 

 the chitinous external layer which covers it may be called 

 after its detachment). The new wing-rudiments are found to 

 lie similarly flat within their sheaths at every change of skin 

 down to and including the last but one, into the interval be- 

 tween which and the last it is that the growth of the wings 

 from small and insignificant rudiments to their full extent is 

 compressed. The penultimate change of skin accomplished, 

 new wing-iiidiments are produced in due course from the 

 cellular layer; and at the time when their sheaths first become 

 detachable from them they, like all their predecessors, lie 

 extended quite flat within these sheaths ; but the detachment of 

 these is no sooner completed than they* commence to grow 

 •with great rapidity. The first outward and visible signs of 

 the growth that now ensues are the thickening of the prolon- 

 gations (which up to this time were thin plates with thin and 

 sharp edges closely embracing the insect's body, but which 

 now gradually become biconvex masses with thick and blunt 

 edges standing out from it) and the gradual obliteration of 

 the principal nervure. The walls of the sheaths eventually 

 become distended to such a high degree of tenuity and conse- 

 quent transparency under the enormous pressure thrown upon 

 them by the rapidly growing wings, that it is possible to see, 

 even without dissection, the manner in which tnese are forced 



* /. €. the winp-rudiments. 



