116 THE DEVELOPMENT OF WINGS 



and when the old intima is molted at the beginning of the fifth stadium the 

 tracheoles become connected with the trachea. At the same time the 

 tracheoles uncoil, and extend in bundles in the forming vein-cavities of the 

 wing-bud (Fig. 115, e). These tracheoles may be termed the larval 

 tracheoles of the wing, as their existence is transient; they supply the 

 developing appendage with air only during the last larval and the prepupal 

 stadia. 



At the beginning of the fifth stadiimi or at the end of the fourth, there 

 begins the development of the tracheee of the wing. These extend out 

 into the developing vein-cavities along with the larval tracheoles. Each 

 is developed as a tubular extension of the epithelium of the large trachea, 

 the mouth of which is closed by the intima of this trachea. For this 

 reason, during the last larval and the prepupal stadia the wing trachea are 

 not functional; they contain no air and are apparently filled with lymph. 

 During this period the wing-bud is supplied with air by the larval tracheoles 

 as indicated above. 



At the molt that marks the beginning of the pupal stadium, the shedding 

 of the old intima of the trachea from which the wing-bud trachea arise, 

 opens the mouths of the wing-tracheae, and they become functional. At 

 the same time, that is early in the pupal stadium, the larval tracheoles 

 degenerate; their function having been assumed by the wing-tracheae. 

 Plate III, Fig. i represents the base of a wing of Bombyx photographed 

 during the prepupal period; the wing-tracheae are still filled with lymph, 

 and the bundles of larv^al tracheoles have begun to degenerate. 



During the pupal stadium a second series of tracheoles are developed. 

 Each of these arises in a cell forming a part of the epithelium of a wing 

 trachea; and they are distributed along the length of these tracheae. At 

 first each tracheole is coiled within the cell in which it originated; but later 

 the tracheoles uncoil and become greatly extended so that the entire wing 

 is traversed by them. Plate III, Fig. 2 is from a photograph of a small 

 part of wing of a pupa of Clisiocampa; the photograph was taken just after 

 the tracheoles had begun to uncoil. This series of tracheoles may be 

 termed the pupal tracheoles of the wing; for they first appear during the 

 pupal stadium. 



As these tracheoles are developed during the pupal stadium within 

 epithelial cells of the wing-tracheee, they can not become functional until 

 their mouths are opened by the shedding of the intima of the wing- tracheae 

 at the final molt. And it is evident that the period of their functioning 

 must be Hmited to the short time during which the wing is expanding and 

 the hypodermis is being used up in the formation of the cuticula of the wing. 

 But as yet no observations have been made to demonstrate this point. 



It seems probable, also, that the supplying of the developing appendage 

 with air by means of the larval tracheoles alone, during the last larval and 



