THE WINGS OF NEUROPTERA 167 



where the coalescence appears to begin. Examples of this condition are to 

 be found in the Berothidae and are discussed later. In Megalomus mcestus, 

 of the Hemerobiidee, vein Sc ends upon vein Ri in the fore wing, while the 

 two end separately in the hind wing. 



Marginal dots or dashes. — While intercalary veins are never developed 

 in the Neuroptera, there are present in the wings of many members of the 

 order small thickened areas alternating with the tips of veins. The areas 

 in some cases are small dots, in others they are slightly elongate. They are 

 probably sense organs, as they bear one or more groups of setae, like those 

 borne by the veins of the wing. These thickened areas are never long 

 enough to be termed veins; for this reason I designate them as marginal 

 dots or dashes. In some cases they are mere dots in one part of the margin 

 of the wing and dashlike in another part. 



The marginal dots and dashes are present quite commonly in the 

 Ithonidas, Berothidae, Polystoechotidae, Psychopsidae, the hemerobiid 

 group of families, Osmylidse, and N^nrnphidae. 



Although they are present in Nymphes, I have not observed them in any 

 other member of the myrmeleonid group of families. 



In some cases they are present throughout the entire margin of the w'ng 

 in other cases they are limited to a portion of the margin. 



The first radio-medial cross-vein of the hind wings. — One of the most 

 characteristic features of the venation of the hind wings of several families 

 of neuropterous insects is the form of the first radio-medial cross-vein, 

 which extends longitudinally in a sigmoid curve instead of transversely 

 (Fig. 162, ist r^m). It arises from the radial sector and extends towards the 

 base of the wing, joining media near its base. 



The fact that this peculiar feature is found in the so-called Megaloptera 

 as well as in the "True Neuroptera" may have some bearing on the pro- 

 posed separation of the Megaloptera from other Neuroptera as a distinct 

 order. 



Frequently there is a fold in the wing which causes the base of the 

 radius to overlap the base of the media, covering the point where this cross- 

 vein joins media. A result of this is that in photographs of such wings the 

 proximal end of the first radio-medial cross-vein appears to be attached to 

 radius. This is the case in the wings of Osmylus hyalinatus represented by 

 Figure 144, which is an accurate copy of a photograph. 



It was probably this condition that led Tillyard to mistake this cross- 

 vein for the base of the radial sector in the Hemerobiidae and to regard the 

 stalk of the radial sector as a cross-vein; this is indicated by his figure of 

 the venation of Hemerohius humtdi (Tillyard '16, p. 285). This matter is 

 of considerable importance as "tJie presence of at least one false or secondary 

 origin for the radial sector in the hind wing" is given as a distinctive charac- 

 teristic of the Hemerobiidae as restricted by him. 



