356 Annals Entomological Society of America [Vol. XII, 



In Prionoxystus robinice, the second branch of the third anal 

 closely parallels the axillary furrow; the third branch follows 

 the free margin of the posterior lobe. As shown later, the inner 

 margin of the fore wing in this species clasps the costa of the 

 hind wing. In other Frenatos, I have found no indication of 

 the presence of the second branch of the third anal vein. It is' 

 suggested as a plausible explanation of the atrophy of this vein 

 that its disappearance is correlated with the loss of the holding 

 function by this part of the wing. 



The cross-vein between the first and second anal veins of 

 the fore wing, which is present in Micropterygidae and Tri- 

 choptera, is preserved in many of the primitive Frenatas as well 

 as in the Hepialidse. It is shown distinctly in Adela hella 

 (Fig. 5), and in Prodoxus; in a pupal wing of Pronuba this vein 

 is as strong as the second anal. It is shown faintly in Cyanauges 

 cyanella. 



In the hind wing, the course of the second anal vein, which 

 anastomoses with the first anal vein for a distance, is regarded 

 by Comstock as a distinctly ordinal character, common to 

 Micropterygidae and Trichoptera (cf. The Wings of Insects, 

 p. 310). The tracheation in the base of the hind wing of 

 Prodoxus (Fig. 3), shows the same course of the second anal. 

 The trachea of the second anal vein bends forward to the first 

 anal vein, but almost immediately bends obliquely backward 

 and after meeting the cross-vein between the second and third 

 anal, extends in a longitudinal direction to the wing margin. 

 The cross-vein is a much more distinct and well developed 

 tubular vein than is the base of the second anal vein itself. 

 The pupal tracheation of Prodoxus (Fig. 6) shows quite clearly 

 the anastomosis of the first and second anal veins. In this 

 instance the longitudinal cross-vein between the second and 

 third anal veins is preceded by a trachea, which is apparently 

 the first branch of the third anal itself. A similar condition 

 is indicated by the course of these veins in Cyanauges and in 

 Adela; in neither is verification possible through persistence of 

 tracheae. In Adela the base of the 2nd A followed proximad, 

 becomes indistinct before reaching the 1st A; the cross-vein is 

 quite distinct. 



In Hepialidae neither venation nor tracheation in the pupal 

 wings (MacGillivray, '12), where there are three free anal 

 tracheae, indicates any such anastomosis. 



