886 IIADIAL AND ZYGOPTEHID SKCTOHS, ifec, 



link, as the trachea moved distad, then the answer is that, in 

 Neosticta, we see a bridge formed per siiltun), and there is no 

 evidence that the trachea did move distad gradually. The effort 

 required to construct evolutionally such an immensely long 

 bridge as that of the Lestidce in this fashion is almost incon- 

 ceivable. Moreover, as a support to the unilateral fork at (), it 

 is absurdly misplaced. One would have expected it to run back 

 to M about half-way between Ms and Mg, if such were the object 

 of its development. 



Further, it should be noted that, in other Orders, the forma- 

 tion of an oblique vein through tracheational specialisation does 

 not involve the formation of a bridge. Thus, in theforewing of 

 the Myymehontidm {i), the vein Mo comes off from Mj by means 

 of an oblique vein, but it is continued basad l)y Cuj, and not by 

 a bridge-vein. In the Odonata, if the bridge-vein be Ms, and 

 not a new development, the position is an analogous one. 



Here I must leave this fascinating line of inquiry, fur which 

 much more evidence would need to be brought forward before 

 it could be regarded as fully established. As regards the main 

 argument, concerning the supposed homology between Ms and 

 Rs, the position may be summed up as follows : — 



1. Trachea Ms is not the homologue of trachea Rs. On this 

 both the structural and ontogenetic evidence aie overwhelming. 



2. If we interpret the main veins solely from their underlying 

 main tracheae, it follows that vein Ms is not the homologue of 

 vein Rs. 



3. If, however, we allow that the venation may retain the 

 archaic plan, while the corresponding tracheation becomes 

 specialised, then we must alter our view of the significance of 

 bridges and oblique veins. In that case, the long bridge of 

 Lestidce may reasonably be regarded as the original vein Ms. 

 In the case of the Anisoptera, the bridge may also represent Ms, 

 while the vein continuing it beyond O is undoubtedly Rs. Thus, 

 the basal portvmsoi the veins now called Ms and Rs respectively 

 would be true homologues ; the distal portions (beyond the 

 oblique vein, when it is present) are never homologous, 



