BY R. J. TILLYARD. 



g'51 



I pointed out a peculiarity in the structure of the media of the 

 Triassic Prohemerobiid Archepsychops (Order Planipennia). In 

 this fossil, it is possible to make out a weak posterior branch of 

 M arising quite close to the base, and running parallel to and 

 between M1.4 and Cuj almost up to the first dichotomy of M1.4. 

 This vein can also be seen in some recent Psychopsidce, as maybe 

 seen from Text-fig.7, a, b. I also figure an interesting formation 

 from thehindwing of the archaic Megalopteron Archichauliodes 

 (Text fig. 7, c). Provided that we are agreed that the arculus is 

 really M 5 , then it is clear that it is the homologue of the vein 

 shown in Text-fig.7. Only, in the Prohemerobiidce and Psycho- 

 psidce, M 6 is in process of degeneration, being crowded out 



Text-fig.7. 

 Portion of base uf wing, to show vestige of M 6 or posterior arculus in the 

 Orders Planipennia and Megaloptera. (a) forewing of Megapsychops 

 Ulidgei (Frogg. ); (6) forewing of Psychopsis elegans (Guer.); (c) hind- 

 wing of Archichatdiodes guttiferus ( Walk. ) ; x specialised elongated 

 cross-vein from M to Rs. (All figures x 12). For rest of lettering, see 

 Text-fig. 2, p. 238. 



between M1.4 and Cu,; whereas, in Archichauliodes, it has suc- 

 ceeded in taking on a form not at all unlike that found in Para- 

 mecoptera, Trichoptera, and Lepidoptera. Whether the con- 

 dition of 3VI 5 in the prohemerobiidce and Psychopsidce is more 

 primitive than that found in Belmontia, it is not very easy to 

 determine But we should note that, in Psychopsis elegans, R 6 

 is frequently found degenerating also, being crowded out between 

 U 4 and M, as in Text-fig.7, b. This seems to point clearly enough 

 to the fact that the condition of M 5 , in those few Planipennia in 

 which it still exists, represents the last stage of degradation, and 



