410 ON THE GENUS CORDULEPHYA, 



men usually associated with the Eucordulina. In this 

 respect, Cordiihjjhya resembles the Idocorautlna. It should 

 be noted, however, that the abdomen of the larger species, 

 (J. montana, is not quite so cylindrical as that of C . pygmcca; 

 so that the narrowing of the body may be merely concurrent 

 with reduction in size. 



iii. The M iny-V cnalwn. — A study of the wing-venation 

 of Cordule/phya (Plate xii., figs. 1-6) seems to shew not the 

 slightest resemblance to the Eucordulina, but, on the other 

 hand, exhibits many close resemblances to the Idocordulina. 

 The impression is, of course, mainly gathered from the region 

 of the triangles. Prof. J. B. Needham* assumes that, in the 

 ancestors of our Atiisoytera, "fore and hind-wings were 

 originally alike," and holds that Cordulcphya is a genus that 

 has perpetuated this zygopterous character. From this basis, 

 he traces the development of the Cordidiimt' by "differentia- 

 tion of the fore and hind-wings, brought about by a number 

 of minor shifts of parts, and chiefly by the broadening of the 

 hind angle of the hind-wing and the development of an anal 

 loop for its support." Now it is evident, since he takes 

 Cordulephya as an illustration of his primitive wing-type, 

 that he would have us infer, that the fore and hindwings of 

 the ancestors of the present-day Corduliin(n were alike in 

 being narrow and of a zyyopterous character, and that the 

 anal loop was a rctnoyenetic development in the Lihellulidoi. 

 Neither the fossil record nor the study of the wings them- 

 selves bears this out. The oldest fossils referable to the 

 I'rofodonata are not zygopterous but anisopterous, though, as 

 might have been expected, there is no triangle or other highly 

 specialised part, such as we associate with present-day An- 

 isoptera. The essential tendency, which, in our Zyyoptera, 

 resulted in enormous reduction of the basal areas, and so led 

 to the petiolate wing, is not indicated in the earliest fossils. 



•Ann. Eiit. Hoc. Ainericd, i., 4, 1908. 



