186 S. II. SCUDDER ON THE WINGS OF SOME FOSSIL NEUROPTERA 



In the Panorpina the wings are extended backwards horizontally or sloping slightly, the 

 lower completely covered by the upper; the opposite pairs divaricate slightly, so as not to 

 overlap one another at all, while the inner edges meet only along the basal half. The anal 

 area is not plicated, nor the costal area deflected. 



In the Phryganina they are extended backwards, steeply sloping from one another 

 obliquely like a roof, at tip generally steeper, nearly vertical, and the opposite pairs ap- 

 pressed ; the anal area of the fore wings deflected horizontally, those of the opposite pairs 

 overlapping one another, the anal area of the hind wings plicated ; the costal area not de- 

 flected. 



In the Hemeristina the wings overlap one another horizontally very completely, even 

 close to the base, probably arched over the abdomen, and probably with the sides protected 

 near the base by the deflected costal area. 



In the Pcdceoptcrina they overlap one another partially in a loose way horizontally over 

 the abdomen, probably with no costal deflection, and in general as in the Termitina, though 

 with not so complete an overlapping. 



Some of the families of Neuroptera will also be found to differ in the position assumed 

 by the wings of the pupa, as follows : — 



Termitina. When developed more than as tubercles they are represented by Westwood 

 as extended horizontally elongate over the abdomen, their inner edges touching at tip. 



Psocina. Horizontally extended backwards in a level plane, not covering much of the 

 abdomen, the posterior covered by the anterior. 



Perlina. Same as in Psocina, but never covering any of the abdomen, the posterior 

 being behind and not overlapped by the anterior. 



Ephcmerina. Curving backwards and upwards over the abdomen, the posterior covered 

 by the anterior, the outer edges meeting along the median line. 



Odonata. Projecting backwards in a level plane, somewhat deflected, the hinder edge 

 downwards, those of either side parallel. 



In the other families the wings are bent over, either curving over upon the breast or 

 extended along the sides, the posterior partially or completely covered by the anterior, or 

 with some modification of one or the other of these modes, so nearly the same as to make 

 the distinctions valueless. 



The families differ also from one another, but agree among themselves in the position of 

 the head. In the Termitina, Embidina, Perlina, Sialina, and Raphidiidae, the head is in the 

 same horizontal plane with the body. In Psocina, Ephemerina, Odonata, Hemerobina, 1 

 Panorpina, Phryganina, Mantispaday and Coniopterygkke, it is in a plane vertical to that of 

 the body. We do not know the position of the head in Hemeristina, but in Pakuopterina 

 it is horizontal. 



From this review of the distinctions among the families we see that the Hemeristina are 

 related to the Hemerobina and Sialina more than to any other, by the mode of branching 

 of the v.scapularis ; to the Ephemerina by the comparative importance of the v. intcrno-media 

 and aiuilis ; to the Odonata by the character of the v. marginalis and v. mediasiina in the basal 

 half of the wing; to the Ephemerina again in the method of dispersion of the cross-veins; 



1 It may be noticed here that the larva of Mvrmelion lias the head horizontal. 



2 This is an additional reason why this should b'j separated as a family group from llaphidiida?. 



