168 DEVELOPMENT OF THE WING-VENATION OF ODOKATA^ 



escaped Needham. Two very important facts need to be empha- 

 sised in connection with it : — 



(l).The so-called "first cubito-anal cross-vein " of Needham is 

 not a true cross- vein at all, but is formed about the anal trachea 

 itself, at the point where it descends from Cu. It may be called 

 the anal crossmgi Ac). 



(2). The so-called " anal vein " of Needham is not a true anal 

 vein, i.e., not developed along the anal trachea, but is a second- 

 aril}^ formed bridge-vein from the lower end of Ac back towards 

 the wing-base, to which it is connected posteriorly in a manner 

 different from the other main veins. Hence the correct notation 

 for this part of the wing is as follows : — 



The cubitus(Cu) of the imaginal wing becomes Cu -f A as far 

 as Ac. 



The "anal vein" (A) must be distinguished as the secondary 

 anal vein, and designated A'. 



To make these points quite clear, let us compare the nymphal 

 and imaginal wings of uEschna hrevistyla Ramb. (Figs. 1-3, and 

 Plate xi., figs. 1-2). The course of the anal trachea is best followed 

 in the hindwing. 



The convergence of A towards Cu near the base, as also that 

 of M towards R, is primarily due to the gradual narrowing of 

 the base of the developing wing in comparison with its length. 

 At first, all these trachecTe are separate; but by the time the wing- 

 cases are half-grown, the fusion of Cu + A, as well as that of 

 R -H M, can be clearly seen. As the wing grows, the fusion 

 becomes greater, until, in the nearly full-grown nymph, A is seen 

 to run obliquely up to join Cu very near its base, and then runs 

 closely alongside it as far as Ac. At this point, A turns sharply 

 downwards away from Cu, in the same manner that M turns 

 away from R at the arculus. Hence Ac is the exact analogue 

 of the upper portion of the arculus. It is, for this reason, that 

 at least one " cubito-anal cross- vein" occurs in every Odonate 

 wing. 



At the lower end of Ac, A branches into two, just as M 

 branches into two when leaving the arculus. These branches 

 were primarily an upper one, A^.g, and a lower one, A4, exactly 



