1913.] NATURAL SCIENCES OF PHILADELPHIA. 255 



If Handlirsch be correct in his conjecture^^ that the Protodonata 

 have descended from such genera of the Dictyoneuridse (Palseodic- 

 tyoptera) as Stenodictya, the dense venation of the former evolved 

 from the equally dense, but more irregular, network of the latter. 

 The Odonata then carried reduction in the network to a greater 

 degree than in the Protodonata and reduction in the rows of cells 

 of the areas now under discussion would, therefore; be a specializa- 

 tion. We must then recognize that as ^arly as the Lias such reduc- 

 tion was carried in Protomyrmeleon to a greater degree than is shown 

 by many living genera and a similar statement must be made for 

 Steleopteron of the Jura. That any of these reduced venations were 

 capable of giving rise to more exuberantly veined descendants would 

 be denied by many phylogenists, but the actual proof of such a 

 denial is not at hand for the Odonata. We have, therefore, only the 

 general tendency to guide us when we assume that reduction in density 

 of venation in the characters of group B means specialization. On 

 the other hand, we must suppose that such richly veined forms 

 of the present day as Thaumatoneura, Calopterijx, Thorc, and Neuro- 

 themis have either preserved the dense venation of, perhaps Mesozoic, 

 ancestors or that they have acquired their density by hypertrophy 

 in the course of generations.*^ Here, as elsewhere in this discussion, 

 the possibility of further light from ontogenetic studies is very 

 evident. 



Summarizing the results of this discussion, we conclude that 

 later phylogenetic status is denoted by the following conditions in 

 the characters of the genera compared in this paper : 



1. Retraction of the nodus toward the wing-base. 



2. Reduction in the number of antenodals. 

 (3. Reduction in the number of postnodals.) 



(7. Reduction in the number of rows of cells l^etween ]\Ii and 



Mia.) 



(8. Reduction in the number of rows of cells between Mm and M2.) 



9. More distal position of origin of M2. 



(10. Reduction in the number of rows of cells between M2 and Rs.) 

 11. More distal position of origin of Rs. 



(12. Reduction in the number of rows of cells between Rs and Ms.) 

 14. More distal position of origin of Ms. 



15a. Reduction in number of cross-veins proximal to the subnodus 

 between Ms and M4. 



8' L.c, p. 305. 



88 The latter alternative is apparently that held by Dr. Ris for the analogous 

 cases of Pantala and Tramea. 1912, p. 46. 



