236 STUDIES IN AUSTRALIAN NRUROPTEKA, iii., 



the wing. Thus Mj, on arching up from mf^ conies at once close 

 up under Rs. There follows an extraordinary adaptation on the 

 part of this latter trachea and its most proximal branch 1^4. 

 By means of an upward curving of Rs below a?, and a bending 

 of B4 convexly to Mj, the triangular area between the thiee 

 trachefe Rs, Mj, and B4 is reduced to a minimum. This area is 

 represented in the imaginal venation by the minute triangle /. 

 Several consequences arise at once from this, which can l)e best 

 noticed by comparing the venation of the hindwing with tliat of 

 the forewing. Firstly, the portion of B4 descending from Rs is 

 almost completely obliterated. Consequently, the place of B4 in 

 bounding the cell marked h is taken by the obliquely ascending 

 portion of Rs, which meets the cross- vein below x. As a further 

 consequence of this arrangement, the cross-vein from x appears 

 to be the real basal part of Rs, and x appears to be the radial 

 fork. It is, indeed, no slur on tlie intelligence of all those who 

 have hitherto worked on this family, that they assumed this to 

 be so without any doubt; for who could possibly have supposed 

 otherwise, unless he had the pupal tracheation before him as 

 convincing proof % A second consequence to be noted is that 

 the true basal portion of Rs, being fused with Mj for almost its 

 whole length, takes on the appearance of being actually the stem 

 of the media; and this appearance is even more definite when we 

 note how closelv the true stem of the media in the forewing 

 resembles this. Thirdly, M, itself, being fused with Rs for such 

 a great part of its length, becomes quite lost in the imaginal 

 venation, except indeed for the small vein ascending from mf to 

 Rs, which is, of course, its extreme basal portion. Fourthly, the 

 true main stem of the media in the hindwing (as will be more 

 fully seen below) comes into line with the pseudo-cubitus, and 

 forms the basal portion of that extraordinarily complex \ ein. 



The course of M, in the hindwing, after it leaves B4, is similar 

 to that in the forewing, and need not be further discussed. M., 

 however, from its origin at mt\ follows a considerably simpler 

 course than it does in the forewing. There is no median arculus, 

 and no median loop. Indeed, M.j simply runs parallel to and 

 below Mj for a considerable distance, finally meeting Mj where it 



