r.Y R. J. TILLYAKD. 



lib 



radius. Sc, R, and Rs did not unite before the tip of the wing. 

 The branches of Rs were numerous, and tended to run down- 

 wards obliquely across the wing, rather than to turn and run 

 parallel to and beneath Rs. M was two-branched, as was also 

 Cu. Both were moderately short, and tended to curve down- 

 wards to the posterior border of the wing, so as to run parallel 

 with the more proximal branches of lis. Three anal s'eins were 

 present, all short. The cross-veins between the branches of Rs 

 tended to become arranged in series, with elimination of those 

 that were unsuitably placed. Text-fig.7 shows a hypothetical 

 ancestral Chrysopid wing conceived on these lines. 



Sc ? T^s 





^,0 ^7 ^' ^^ 



Text-fig.7. — Hj-^pothetical ancestral wing-fonn of tlie Chrysopidcp. 

 Paloeontology here comes in, and affords us valuable proof 

 that our theory of Chrysopid descent has a solid basis of fact. 

 Text-fig.8 shows the wing of }fesochrysopa zitteli Meunier, from 

 the Lithographic Chalk of Bavaria (Upper Jurassic). This wing 

 was originally described by Meunier as a Termite (genus Hagenio- 

 termes), hut was i-e-studied and figured by Handlirsch ("Die 

 Fossilen Insekten," p.612, and Plate xlviii., fig. 14). This author 

 proposed the generic name Mesochrysopa for it, and placed it as 

 the unique representative of a new family, Jlesochrysopidte. His 

 comment on the character of this family is very instructive, if 

 read in the light of our new knowledge of Chrysopid venation. 

 He says :^" Diese Gruppe bildet zweifellos ein Bindeglied 

 zwischen den Prohemerobiden und Chrysopiden. JViiterstei'en 

 hat sie den noch ungebrochenen Radius gemein," (here he un- 

 doubtedly intended to say " Radial sector" for "Radius"); " mit 

 letzteren die Bildung des Cubitus und der Analadern, sowie die 



