XVI] THE GEOLOGICAL RECORD 305 



together. This condition is unparalleled amongst recent forms. 

 The other interesting points about the wings of this insect are : 



(1) The presence of an unveined membranous precostal area, 

 not found in any recent Dragonfly, (2) the passage of Sc almost 

 to the extreme tip of the wing, (3) the slight separation of M 

 basally as a fine vein just below R, foreshadowing the complete 

 fusion seen in recent forms, (4) an undulation of Cu and A near 

 the base, (5) the regular arrangement of the interpolated sectors 

 and cross-veins. 



In Meganeurula selysi, there is a further marked advance 

 towards the true Odonate type by the fusion of Sc with the 

 anterior border of the wing, at a point less than half the total 

 distance from base to tip. This foreshadows the formation of 

 a true nodus. The radius, too, runs into the same margin at 

 about the region of the more modern pterostigma, thus suggesting 

 the origin of that organ as a very long and narrow membranous 

 thickening between C and R at this position. One of the most 

 archaic of surviving Anisopterid subfamilies, the Petalurinae 

 (fig. 18 B), has a pterostigma not far removed from this form. 



The recent discovery by Bolton [ii] of portion of a Meganeurid 

 wing (M. radstockensis) in the Forest of Dean coal-measures in 

 England, throws further light on the genus. Only parts of the 

 basal half are preserved, but we can note two new peculiarities. 

 The precostal area is absent, but the costa itself is much thickened 

 basally, and has a coriaceous border. A remnant of this condition 

 still obtains in Petalura. Further distad, the costa is ribbed in 

 the regular manner seen in recent Dragonflies. More interesting 

 still, the slanting cross- vein A c, recently shewn by me[i76] to indicate 

 the point where the true anal vein diverges from Cu, is clearly 

 to be seen. Hence it would appear that Cu and A were fused 

 basally, as in all recent forms. Thus the gap between Protodonata 

 and Odonata is being gradually lessened, until to-day we may 

 almost certainly see in the Meganeuridae the giant relatives of 

 the direct ancestors of some at least of our recent families. 



Only two other families of Protodonata are known, each 

 represented by a single genus and species. The Paralogidae 

 (Paralogus aeschno'ides Scudder) had a rather short and broad 

 wing, with Sc ending about half-way, and R -f M completely 



T. D.-F. 20 



