222 WING-VENATION IN ZYGOPTEROUS DRAGONPLIES, 



On the totality of the above evidence, which I consider to be 

 very strong, a new view of the venation of Zygoptera seems to 

 me to be not only indicated, but absolutely necessary. For this 

 new view, which I now present for the first time, no better name 

 than the " Unbranched Radius Theory" (the name given by Dr. 

 Ris) can be found. We may enunciate the theory as follows. 



The basic difference between Anisoptera and Zygoptera is the 

 condition of the radius, a difference which is more radical than 

 that of the presence or absence of the triangle, since it goes 

 further back in point of geological time. In all Anisoptera, the 

 radius has a single branch, the radial sector (Rs), which crosses 

 two branches of the media (Mj.a); a short bridge is also developed 

 in connection with it. In all Zygoptera, the radius is unbranched; 

 a bridge is absent,* except in Lestidce, when it is very long. In 

 A7iisopte7'a, the media (M) gives off only th7'ee posterior branches, 

 so that Rs lies between the second and third, counting from the 

 base. In Zygoptera, the media gives off /our posterior branches, 

 one of which takes the place of Rs, i.e., supplies (in the larval 

 wing) the area lying between the second and what is here the 

 fourth branch of M, counting from the base. 



In the Protodonata, the condition of the radius was variable, 

 and of no greater than generic value. Fixation of the condition 

 was most probably correlated with the formation of arculus, 

 nodus, and pterostigma, which began in the Permian. Differ- 

 ence in wing-breadth may have been the original determinant 

 whether Rs was to be present or not. In any case, intermediate 

 forms died out before the Lias was reached. We there find the 

 Odonata represented by a race of robust forms — the Anisoptera — 

 with the radial sector present, and the discoidal cell divided into 

 triangle + supertriangle; and by a race of slenderer forms (many 

 of them still of large size) — the Zygoptera — with the radial sector 

 absent, and the discoidal cell not definitely divided. 



The development of the three main types of Zygopterid wing- 

 venation may be shown thus : — A glance at Fig. 3, A, B, shows 

 that, on the ontogenetic evidence, the Calopterygidce have tended 



* There may be a very short bridge in Rhinocypha and Ghalcopteryx, but 

 the larvae are not available for examination. 



