BY R. J. TILLYARD. 297 



in an otherwise pigmented wing. For instance, in Psychopsis 

 UUdyei, the whole of the outer gradate series runs through a 

 deeply pigmented brown mark on the 

 wing; but each separate cross- vein is 

 enclosed in a tiny clear area of the 

 kind shown in Text fig. 7, a. If we 

 admit that the fusion of our cross- 

 veins, in the case of Drepanepteryx\ 

 took place originally as a specialisa- 

 tion for mechanical advantage, and that, at the same time, these 

 cross-veins happened to be enclosed in small, clear areas as they 

 are in Psychopsis illidgei, we get at once a weak fenestella of the 

 form seen in D. hii.tnilis. A slight extension of this gives us the 

 well-formed fenestella of D. phalcendides. Thus it would seem 

 that the formation of this peculiar structure may well be due to 

 the accumulation of small, advantageous variations to the benefit 

 of the species, and that it is really an excellent illusti-ation of the 

 action of "natural selection" in the strict. Darwinian sense. 



In Drepanacra, the forewing is considerably less falcate than 

 it is in Drepaneptery.):. It is interesting, therefore, to find in 

 this genus the development of still another structure of the same 

 nature as the fenestella, \iz., a set of one or more, white (unpig- 

 mented) lunides along the falcate border of the wing. One of 

 these lunules is shown in Plate xiii., fig. 14, enlarged, Avhile the 

 set of three, usually present in D. humilis, is shown in fig. 12 of 

 the same Plate. These structures not only serve to increase the 

 appareiit falcafioit of the wing, but, when well developed, they 

 sugge-st a series of small "bites" or " tears"' out of the edge of 

 the dead leaf, which the wing so closely repi-esents in repose, and 

 thus contribute, presumably, to the innnunity from attack enjoyed 

 by these archaic insects. 



As regards the, frenulum of Drepmmrra (Plate xi\-., fig. 19), it 

 is of a more specialised form than that figured by Sharp(13; p. 168) 

 for Drepaneptery. If. I find, in the males of D. humilis, that the 



"' Diagrams to .show the evolution of a fenestella : a, gradate crossvein.s 

 in small, clear areas; b, alignment of .same; c, fenestella completed. 



