2. KOCTUID.E. 



usually evenly curved, often crenulate, rarely produced to points, 

 or angled, the inner margin sometimes highly lobed ; vein 1 a 

 weak, not anastomosing with ] 6 ; 1 c absent ; veins 3 and 5 

 usually from near angle of cell, but 3 sometimes from well before 

 angle ; 6 from upper angle ; 9 from 10 anastomosing with 8 to 

 form the areole ; 1 1 from cell ; in Cyligramma 7, 8, 9 stalked ; 10, 

 11 fi'om cell much attenuated and approximated to the costa. Hind 

 wing with the termen usually evenly curved, rarely angled, in 

 Pterocyclopliora produced to points and a short tail ; veins 1 a and 

 1 h present ; 1 c absent; the cell usually about half the length of 

 wing, but often about one-third length of wing, in Anisoneura 

 about one-fourth, and in Agonista and Nyctijjcio about one-fifth of 

 wing ; veins 3 and 4 from angle of cell, 5 fully develoj^ed from or 

 from just above angle; 6, 7 from upper angle; 8 anastomosing 

 with the cell near base, then diverging, in AcantliodeUa and 

 Dory odes approximated to the cell to middle ; in the males of some 

 species of Nyctipao veins 3 or 4 run to the functional apex of the 

 hind wing whilst veins 4 or 5 to 8 are aborted in a lobe at base of 

 costa containing a glandular patch of scales, the lobe being the 

 reduced costal half of the wing ; the wings in the males are some- 

 times more or less entirely clothed by patches of androconia or 

 woolly hair above or below, the hind wing below sometimes with 

 ridges of hair on the veins, or a fold and fringe of hair on inner 

 area. 



The spines of the hind tibiae are not a very constant character, 

 and too great reliance should not be placed on it, besides which 

 the spines are often difficult to see especially in the males with 

 hairy tibiae and are easily broken off. 



The Catocalince form a well-characterised and natural subfamily, 

 but many of its most divergent forms such as PhyUodes, Cocytodes, 

 JSypnoides, Eccrita, Attatlia, Opliisma, Leucanitis, Hypogrammodes, 

 and Pericyma have extremely closely allied forms in the Nnctuincp, 

 and it seems probable either that the various genera are derived 



Fig. 1. — Lai-va of Caioccda fraxini. \. 



from different parts of the Noctuince or that spiny tibiae are an 

 ancestral character in the Noctuidce, and the Noctu%n<x are derived 

 from various branches of the Catocalince ; in either case they would 

 have to be united as one very large subfamily, or divided into sub- 

 families or other characters. 



In the Phytometrina' (Phisianci;) some of the genera have the tibiae 



