12 Journal New York Entomological Society. [Vol. vi. 



main directions Endromis shows the most progress. Still, we shall have 

 to discuss the relation between these specializations and habit, although 

 here the matter may detain us no further. 



It may be here remarked that it is not strictly correct to speak of 

 the Cubitus "becoming three or four branched.'' The Cubitus is always 

 two-branched. It is the movement of the lower branches of the 

 Media, which become varyingly attached to the Cubitus, thereby giving 

 the appearance of increasing the number of the cubital veins. In the 

 opposite direction, it is the same way with the Radius. 



sphingim:. 



The absence of homology between the anal horn, of the Hawk Moths 

 and the similarly situated hypertrophied tubercle of the Emperor Moths 

 has been determined by Dyar and is illustrated by me in the " Satur- 

 niiden,".pp. 7-8. The two groups have then no immediate connection 

 ana the correspondence with the Citheronians is illusory, the common 

 habit of pupation of secondary acquirement. The venation, both of 

 Endromis and Sphinx, entirely warrants this view of the case. We 

 need not detain ourselves with these matters here but pass on to the 

 venation. The mass of preparations I have made show me that this offers 

 no characters of precision for a division into subfamilies. The wings 

 appear cast, like iron, into the same mould. Still there is a play with 

 the branches of the Media and it is often not difficult to decide, as be- 

 tween distinct forms, which is the more specialized. Harder to embrace 

 these forms into groups. A form like Cephonodes picus seems special- 

 ized from the amount of absorption of vein IVi by the Radius on 

 primaries, the retreating, almost vanished cell on secondaries, the fusion 

 of IV3 with Vi. Cephonodes is more specialized than Hemaris. As 

 between Macroglosstim and Aellopos it is hard to distinguish ; they seem 

 practically identical. The obliquely transverse and rigid crossvein of 

 primaries is the same and all goes to show that the position assigned by 

 me to Aellopos in 1865, among the Macroglossians, is correct and that 

 its placement among the Chserocampians in the Philadelphia List is er- 

 roneous. A study of the neuration seems to favor the idea that the 

 Macroglossians are really the more highly specialized of all the groups. 

 On the other hand, that portion of the hind wing between Cubitus and 

 the anal margin appears generally more lappet-like in the Macroglossians 

 (shared by Aellopos') as compared with the Elephant Hawk Moths. 

 There is a decided indentation of the outer margin between V2 and VII. 

 Almost does this character seem a probable test to distinguish the groups. 



