618 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.48. 



species involved large, and were such possible new genera not 

 monotypic, added convenience might permit the erection of such 

 new genera; at the present time there seems no warrant for such a 

 minute breaking up of the genus. 



The four genera recognized in the present study are described in 

 the following key and tabulation of venational characters. They are 

 as follows : 



1. Protoneura; type, capillaris Rambur. 



2. Epipleoneura, new genus; type, lamina, new species. 



3. Psaironeura, new genus; type, remissa Calvert. 



4. Epipotoneura, new genus; type, nehalennia, new species. 

 The new generic names are suggestive of the manner of flight of 



these insects. In their compounding and in the selection of specific 

 names for the six new species described in this paper I have had the 

 advice of Prof. J. B. Parker. 



In agrionine wings generally, and in those with reduced venation 

 especially, any shifting of one of the larger veins results in other 

 striking readjustments of other parts. Consequently in such wings 

 as Protoneura calverti, for example, where the maximum reduction 

 of American agrionines is reached, there is a relative fixity of parts, 

 which in a more complex wing might shift position with but little 

 effect on adjoining parts. For example, the postnodals generally 

 have well-defined descending cross veins with one of which M 2 bears 

 a definite relation. 1 The origin of M 2 is thus, when once fixed, 

 pretty rigidly held in its place. When one part or character of a 

 wing is located or described by reference to another part, and varia- 

 tion appears, it may be impossible or difficult to tell which part has 

 shifted. For example, variation in the position of the cubito-anal 

 cross vein relative to the first antenodal may result from the shifting 

 of either one or both of the two veins. A series of wing photographs 

 of a species, enlarged to the same scale, would enable one to deter- 

 mine the definiteness or instability of the position in the wing of the 

 cubito-anal cross vein, the antenodals, the arculus, the nodus, and 

 other parts. Such a study should be made in the interests of 

 accuracy in defining the position of these parts. In tabulating wings 

 I have noticed some shifting of the cubito-anal cross vein and both 

 antenodals, and have detected no instability in the arculus or nodus. 

 In one species it was noticed that, when the second antenodal is 

 close to the arculus, the upper limb of the arculus is shorter than 

 when the antenodal is more proximal. 



i Compare the definite and normal bracing of these parts in calverti (fig. 1), for example, with the weak 

 and unusual arrangement shown in the front wing of remissa (fig. 9), pis. 41 and 42. 



