200 DEVELOPMENT OF THE WING-VENATION OF ODONATA, 



and hence gets a plentiful supply of oxygen, and grows to its 

 full length. In tiiat case, we must transpose the lettering in 

 the plate, and name the more proximal trachea Ks. 



If we accept the above solution, the ProtoiiPAirini (so far as 

 known) will differ from all other Agrioiiidce in having Rs crossed 

 under three branches of the nodus instead of two. The question, 

 however, should not be regarded as finally settled, since it is only 

 fair to point out that the wing of Neosticta is so highly pigmented, 

 and the trachea Rs so faintly indicated, that I may have made a 

 mistake in my interpretation of its course. In Plate xiii., fig. 4, 

 I give the actual photograph itself, but I may add that the wing 

 itself, Avhen first examined, was somewhat clearer than the photo- 

 graph, and gave me a distinct impression of Rs crossing under 

 Mo as well as Mj.o. The point can only be finally settled by 

 further investigations, if possible on Jarvse fresh from an ecdysis, 

 at a time when pigmentation is less dense. 



Concerning the rest of the Zygopterid nymphal wing, there is 

 very little of interest. The arculus is formed exactly as in 

 Anisoptera by M descending from R, but the supporting vein 

 formed underneath it, and completing it, also forms, of course, 

 the basal or proximal side of the quadrilateral. The distal side 

 of this latter is another cross-vein connecting Mj with Cu at its 

 point of bifurcation. The subcosta is a very weak trachea, and 

 takes very little part in forming the nodus, which is here con- 

 structed from a small thick trachea arising from R, but princi- 

 pally by a simple vein, devoid of trache?e, and blocked off by a 

 huge mass of pigment proximally. This formation is an astheno- 

 genetic specialisation derived from the older method shown in 

 Anisojitera (where Sc plays the principal part), and is due to the 

 weakening of tSc, caused by the narrowing of the wing. 



All the wings studied under the tribe Agrionini agree with 

 Galiagrion in all essential particulars, except that Ac and its 

 branches are not so clearly to be seen as in the larger larva. The 

 distribution of trachea? in Argiolestes is also essentially similar to 

 that in Caliagrion, the chief difference being that, in the latter, 

 all trachcc^ run nearl}^ parallel or slightly converging, while in 

 the former, the wing being broader apically, they tend to diverge 



