24 CONNECTICUT GEOL. AND NAT. HIST. SURVEY [Bull. 



veins became fewer in number, some were lost, and forms appeared 

 with venation very similar to those existing to-day. Kennedya 

 mirahilis Till., the oldest known dragonfly (recently discovered 

 in the Permian of Kansas), has a wing very similar to the Zygop- 

 tera of today except that it is narrower and there are fewer 

 cross-veins. 



Three general tendencies in wing-vein reduction during the 

 period following the Carboniferous may be noted: (i) Reduction 

 in the costal area — the costal and anterior radial branches were 

 much branched in Carboniferous forms. In later forms there is 

 little branching of these veins. (2) Reduction in the number of 

 primary veins in the anal area, and (3) the loss of one or two 

 veins in the center of the wing by the crowding together of the 

 wing at the base. 



As already mentioned, two or more systems have been advanced 

 to account for the curious venation of the dragonfly wing. That 

 of Tillyard is in reality a modification of Lameere's system. Both 

 Lameere's and Tillyard's works are the outcome of the work of 

 Comstock and Needham, although important changes have been 

 made necessary by the fossil discoveries mentioned. Tillyard's 

 system is based primarily upon the presence of convex and con- 

 cave wing veins which he is able to trace back through a series of 

 forms to the Paleodictyoptera and Carboniferous Protodonata. 

 It provides for six convex or plus veins and five concave or minus 

 veins, the theory being that simple branches or bifurcations of 

 the originals retain the same sign, and that veins appearing 

 between these bear the opposite sign. Thus should Radius become 

 two-branched, each branch would have the original sign, let us say 

 plus (convex). Now should a third branch appear between these 

 two and become connected at the base with either, it would bear 

 the opposite sign, or in other words, would be a concave vein. 

 One difficulty appears to lie in the fusion of veins and the resulting 

 sign which the fused vein should receive — and it seems apparent 

 that more work is needed, especially in the region of the triangle 

 where the veins have been so profoundly altered. It is quite evi- 

 dent that Cui has disappeared from the picture (see Fig. 4, a, b) 

 and it is probable that the posterior branch of M has also disap- 

 peared. It should be noted further that the Rs of Needham is 

 not the true Rs of other orders but is a recent development and 

 in reality a minor offshoot of R^. The system of Tillyard and 

 Lameere probably corresponds more fully with the true develop- 

 ment of Odonata from the early geological periods, but since the 

 system of Comstock is at present more widely known and gen- 

 erally used, and since all descriptions were drawn up before the 

 advancement of Tillyard's theories, it is used in this paper. There 

 will, no doubt, be a gradual change to the newer nomenclature as 



