NORTH AMERICAN LEPIDOPTERA. 151 



There has been some difficulty in dividing this subfamily. A few 

 of the genera there is no ti'ouble with, such as Amphonyx, Dlhpho- 

 nofii and Caidethia, and these I place at the head of the series. 



Amphonyx contains one of the largest of our American sjjecies, 

 and the genus is easily recognizable l)y the prominent head, large, 

 unlashed eyes, untufted thorax and the j^eculiar naked hooked ter- 

 minal joint of palpi, which look like a pair of small horns out of 

 the vestiture of the head. 



Dllophouota shares this latter peculiarity in some species, in a 

 much less marked degree, but is evidently distinct by the thoracic 

 divided crest, which is unitjue in the sulifamily. 



Cantethia contains a single small sjjccies with prominent head and 

 eyes, untufted thorax and body, and unarmed legs. The secondaries 

 are reddish or orange. 



The remainder of the genera offer no salient characters, except in 

 Elleina, which has the tongue obsolete. The chai-acters here used 

 are primarially the lashed or unlashed eyes, an extremely uncertain 

 character ; the armature of the legs, which here is unreliable and 

 very variable, and some slight differences in the wing form. A ref- 

 erence of all the species to the genus Sphinx would not l)e indefen- 

 sible, and it becomes a mere question of the desirability of subdivi- 

 sion. Prof. Fernald places S. plebeius in Hyloicus, and yet cannot 

 quite satisfy himself that it is not congeneric with Phlegethontius 

 Species not seen by Prof Fernald, however, make a connection with 

 Sphuix complete, while the only things separating Ceratomia from 

 Hyloicus are the unlashed eyes and slightly retracted head, charac- 

 ters in which cupressi forms almost the exact intermediate. It is, 

 therefore, nothing at all but a question of convenience. The order 

 of arrangement is similarly complicated by forms like hykt'us, which 

 are somewhat abei'rant wherever placed. After due considei-ation 

 the following scheme seems not unnatural : 



First in the series might come Protoparce, with rather distinct or 

 prominent head, unlashed eyes, very long tongue, sparsely or non- 

 spinose fore tibite and armed fore tarsi. 



To this Sphinx will join naturally. The eyes are smaller, head 

 more retracted, eyes lashed, anterior tibise always, middle tibiae gen- 

 erally, spinose, fore tarsi variable as to armature. 



Dolba is s|)hingif()rm in aj)pearance and has similarly lashed eyes, 

 differing in the entirely non-spinose legs. It bears a superficial i-c- 

 semblance in ornamentation to Protoparce rustlcu, l)ut is very iiiuch 

 smaller. 



