022 THE PANORPOID COMPLEX, iii., 



M 3+4 became rapidly obliterated, leaving only the short basal 

 connection from mf' to m-cu, and a short distal stump-vein pro- 

 jecting inwards from the point of origin of M 3 and M 4 . Also, 

 the distal free piece of M 3 became rapidly obliterated, leaving 

 only a short stump attached to im, and closely resembling the 

 stump of M 2 above it. Naturally, I expected that these stumps 

 would remain, when the wing had become quite hardened up. 

 But I was surprised to find that both the outwardly projecting 

 stump of M 3 , and the inwardly projecting stump of M 3 , 4> 

 became finally eliminated, leaving a single unbroken and only 

 slightly waved vein connecting C 1 u x below witn M 1+2 above. 

 Comstock (15, p. 356) has, naturally enough, taken this vein to 

 be the inter-median cross- vein (m of his notation, im of mine) ; 

 whereas it is now seen to be a composite vein, made up of a 

 short piece, im, above, a long piece of M 3 in the middle, and a 

 short piece of M 4 below. Another change in the hardened 

 wing was the straightening out of the basal stump of M 3+4 , to- 

 gether with the cross-vein m-cu, to form a single strong trans- 

 verse brace-vein, which might well be named the secondary 

 arculus, seeing that it plays the part of bracing the wing in the 

 same region as the original arculus does in older types. Care 

 must be taken to distinguish the two formations, as there is no 

 trace left of the original cubito-median Y-vein in the Myiodaria. 



(4) The course of Cu 2| and the line of the anal furrow («/) 

 running below it, are visible in the freshly expanded wing, but 

 are obliterated later. 1A, part of 2 A, and the cross-vein cu-a 

 remain chitinised in the hardened wing. 



The above evidence seems to me tp be of very great value, as 

 indicating a new method of studying the venation of highly 

 specialised types of Diptera. In particular, I think that every 

 occurrence of stump-veins ought to be noted down, and attempts 

 made to find out what their condition is in the freshly expanded 

 imaginal wing. In this connection I will now only refer to the 

 stump-vein arising from R 4 in many Tabanidae (Text-fig. 68). 

 I have taken this as the remnant of the inter-radial cross-vein 

 ir, which is its most obvious interpretation. But I hope that 

 any entomologist, who may have the opportunity, will examine 

 the freshly formed wing of any suitable species, to see what 

 other formations may be indicated there (as, for instance, the 

 possible presence of a separate R 3 ) . 



It should be noted here that there is a very great difference 



