642 THE PANORPOID COMPLEX, iii., 



to be evident, a fairly strong trachea was seen traversing M 5 , in 

 the forewing; but this trachea was not of so large a calibre as the 

 cubital trachea, which still ran strongly in the lower arm of the 

 Y-vein. There is no trachea in M 4 . Trachea Cu 2 is still fur- 

 ther reduced in the forewing. The anal area has become very 

 much narrowed, and tracheae 2A and 3A are very much reduced 

 in the forewing. In the hindwing, the fusion of Cu 2 with 1A 

 remains, but has approached nearer to the base of the wing, as 

 has also the position of M 5 in both wings. This points to a 

 great contraction of the base of the wing in the late pupa, and 

 this contraction is very much further emphasised at metamor- 

 phosis, when the more distal part of the wing becomes much 

 expanded. 



It is necessary here to point out that trachea M 5 does not 

 always appear in the late pupa. In some cases, it may be seen 

 as a very small trachea indeed; while, in one pupa of Leto, ex- 

 amined only a few days before metamorphosis, there was still 

 no sign of it. 



If a freshly-emerged Hepialid imago be taken and killed, and 

 its wings carefully descaled and cleared, the tracheae may be 

 found in situ, running in the veins, very much as they are to 

 be seen in the late pupa. The principal changes that take place 

 at metamorphosis are as follows : — 



(1) The base of the wing, already reduced in size in late 

 pupal life, undergoes practically no expansion. On the other 

 hand, the more distal portion expands greatly, so that the final 

 result is to bring the primary median and cubital forks very 

 much closer to the base of the wing than they were in the pupa. 



(2) Trachea M 5 is frequently of large calibre, and may be 

 considerably larger than trachea Cu 1# There is, however, a great 

 deal of variability in this character, as I have examined a num- 

 ber of imagines in which trachea Cuj was considerably the 

 stronger of the two. 



(3) Trachea M 4 is usually absent; but it was present in a fair 

 number of wings, and in some cases it was even of larger calibre 

 than trachea Cu la< In a few specimens, I have seen the latter 

 quite aborted; but in most cases, when present, the two tracheae 

 run side by side along the stem of the distal Y-vein, M 4 -f Cu la , 

 (For details, see Section vi.) 



(4) In the case of Ch. splendens (Text-fig. 78) scarcely any 

 trace of either 2A or 3A is left in the forewing of the imago, 



