1917] Pupal Wings of Nepticula 235 



for merely a short distance into the base of the costal vein, not 

 reaching the humeral cross-vein. In N. rosaefoliella (Fig. 3) 

 no costal trachea could be distinguished. 



Subcosta. — The subcostal trachea is distinct and branched 

 near its tip in N. platanella; Sci (except early as shown in Fig. 2) 

 is shriveled and curled even in the pupa, while Sc2 passes straight 

 onward to the margin of the wing. There is no indication of 

 branching in the subcostal vein of the imago in which trachea 

 Sco extends to the end; the shriveled stump of Sci occasionally 

 persists within the subcostal vein. In A^. rosaefoliella, the 

 subcostal trachea is unbranched, that portion persisting being 

 Sc2. The humeral cross-vein is easily visible in N. platanella, 

 and I have found it present in all specimens examined ; it passes 

 obliquely from subcosta to costa near the base. In the imago 

 it arises almost at the base of the wing, so oblique as almost 

 to seem a proximal prolongation of costa. 



Radius. — With the exception that R2+3 is represented by an 

 unbranched trachea, the tracheation in N. platanella approaches 

 closely to that of the hypothetical type. The tracheal stem of 

 radius branches dichotomously, about half way between the 

 base of the wing and the point of separation of vein Ri from 

 the main radial stem, into an unbranched trachea Ri, and into 

 a stem which again divides dichotomously, also before the 

 separation of vein Ri, into a second unbranched trachea, 

 Ro+3 and into R4-^5, so that for a short distance the developing 

 vein cavity contains three tracheal branches lying closely along 

 side of one another. R24-3 follows R4+5 for a short distance 

 before diverging toward the costa. A little beyond the point 

 of separation of R2+3, the trachea of R4+5 divides, the branches 

 however lying close together in the same vein cavity until near 

 the apex of the wing, where they diverge, each lying in a separate 

 vein cavity. The relation of tracheae to veins in N. rosaefoliella 

 is in essential agreement with that observed in A^. platanella, 

 except that (1) the branching off of tracheae Ri and R2+3 occurs 

 at or near the point of separation of the corresponding veins 

 and (2) there is no division of trachea R44.5 until very near the 

 margin of the wing (cf. fig. 3) and but a single unbranched vein 

 is formed. 



In the imaginal wing of the European Trifurcula, the main 

 stem of the persistent radial trachea divides dichotomously into 



