NOCTUINA. i8i 



outer edge of the discal cell. In the vast majority of cases 

 in which vein 5 is moderately or extremely thin and weak, 

 and produces little impression upon the margin, it arises 

 from the middle of this cross-bar. When it arises upon or 

 near the median nervure it is usually strong and produces 

 the marginal extension. As before remarked, the main fold 

 of the wing by which it is laid together and packed under 

 the narrow fore wing is down this vein 5, and very often the 

 fold is so strong and obvious that the nervure seems to arise 

 from the base of the wing and pass down the middle of the 

 discal cell. The other fold, when in use, is between veins 1 

 and 2. A curious peculiarity of many species which seem to 

 have the median nervure quadrifid is that it is actually only 

 bifid, vein 5 arising at a very short distance from the median, 

 and vein 3 crossing that nervure to meet it. This, however, 

 is not very noticeable until the wing is denuded of scales. 



The structural character, however, which seems to work 

 most havoc in the arrangement in question is furnished by 

 the eyes. In the majority of species the surface of the eyes 

 is, as in the Bomhyces, smooth and naked, though often the 

 corneous surface is so shrunken after death as to present a 

 more or less irregular pattern of black blotches. But in a 

 less number of species the whole large convex surface of the 

 eyes is protected by short and minute, but perfectly even 

 and regular, upstanding hairs or bristles. These are so ex- 

 quisitely fine that they only become visible under a good lens, 

 and even then cannot be seen from above ; but when looked at 

 across the surface of the eye present a most delicate and 

 regular cloud of tiny short points, perfectly erect and usually 

 of a bright brown colour. In this respect no intermediate 

 gradations seem to occur ; all are either completely hairy — or 

 naked — eyed. Yet to adopt this character alone for separation 

 of the Noetidna into two families would land the systematist 

 into greater difficulties, since this curious character cuts 

 right across some genera which in other respects appear to be 

 tolerably homogeneous. Thus, the genus Cymatophora, as 



