BY B. J. TILL YARD. 563 



scribed and figured (26) . There is, however, no precedent 

 trachea in M 5 in the pupal wing, and the main stem of the Y 

 is supplied only by the trachea Cu'i. 



The same condition is to be found in the pupal wings of 

 Hepialidae, of which I have dissected out a large number for 

 the purposes of this paper. Pupae of the following genera have 

 been dissected: Leto, Ckaragia, Porina and Oncopera, the two 

 latter being supplied by Mr. G. Lyell of Gisborne, Vic. In 

 all of these the condition of the Y-vein is the same, viz. M 5 is 

 represented by a pale band without any underlying trachea, while 

 the trachea supplying the main stem of the Y is derived 

 from Cuj.. (Text-figs. 77, 79.) 



The only pupal wings in which any sign of trachea M 5 can 

 be detected are those of Xyleutes (fam. Cossidae). In two of 

 the five pupa? which I dissected, a short but stout branch of M 

 was to be seen arising near the base, and proceeding distad, 

 below the main stem of M, towards the pale band which repre- 

 sents the upper arm of the Y, and which, in this genus, is 

 transversely placed, resembling a cross-vein. It is possible that 

 this trachea may be the lost trachea belonging to M 5 . 



In advanced pupae, I have sometimes found a fairly large 

 trachea in M 5 ; but this is exceptional, as far as my experience 

 goes. In no case have I seen this trachea penetrate far along 

 the main stem of the Y-vein. 



The conditions in the freshly emerged imago, of which I have 

 examined a very large number, are very different from those 

 of the pupa. A very strong trachea arises from M, and, pass- 

 ing down M 5 , supplies the main stem of the Y-vein right to 

 its distal ends. The original cubital trachea which is found 

 in this vein in the pupa, is often completely ousted, and its 

 truncated remains may frequently be seen near the base of the 

 wing, as in the case of Wingia (fam. Oecophoridae, Text-figs. 

 91, 92) . It is, therefore, necessary to conclude that the main 

 change takes place at, or very near, metamorphosis, at the 

 same time that M 5 becomes a strongly chitinised vein, and the 

 base of Cu becomes almost completely aborted. 



It should also be noted that, in the Hepialidae, the condition 

 of the tracheation in the freshly formed imaginal wing is very 

 variable. Comstock (15, p. 328) states that, in all the wings 

 of this family examined by him, M 5 (which he calls the pos- 

 terior arculus) "is traversed by a branch of the median trachea, 



