-7 



THE CANADIAN ENTOMOLOGIST. 343 



Ennomido;. 

 I. Hind wings, 8 coalescing with cell at base Fernaldellinje. 



Hind wings, S separate from cell 2. 



Frenulum absent Palyadinae. 



Frenulum present 3. 



3. Antennai nearly joined at base Sphecetodinpe. 



Antennpe normally placed 4. 



4. Antenn.ne three-fourths length of fore wings ; legs very 



long Mecoceratinte. 



Antennte never more than two-thirds length of fore wing; legs 



normal 5. 



5. Fore wings, 7 separate from 8 and 9 Melanchroiinse 



Fore wings, 7 stemmed with 8 and 9 Ennominpe. 



From this group is excluded the sub-family of Strophidiinse, now 

 constituting the family Epiplemidae. They are sufificiently removed from 

 the typical forms of Geometrid?e to warrant this perhaps, by the want of 

 humeral angle at the base of vein 9 on the hind wings, and, in the case 

 of Callizzia, by the broadening of the inner margin of the hind wings, 

 which, when the insect is at rest, is rolled round the body, in the manner 

 of the Tineoidea, the upper half overlapping it, while the fore wings are 

 extended at right angles, as in most Geometridae. 



Starting with the sub-family of Hydriominte, I have amalgamated with 

 it the Dyspteridinae, a sub-family established by Dr. Hulst, upon the 

 supposed absence of the frenulum, in certain species. Finding that this 

 appears to be constant in only two of his species, and that in the 

 F.nnomidi^ the same feature occurs, I have determined to abandon its 

 use, as showing when absent merely a tendency toward degeneration 

 (as in the Sterrhinie and Geometrin;x:?, by the partial obsolescence of the 

 hind legs), and not worthy of basic consideration. 



The Brephinae find a final resting place at the close of the family of 

 Geometridfe, for by the presence of vein 5 in the hind wings they belong 

 with this series, and the presence of hair pencil on hind tibiae of c? in 

 Brephos infajis (an observation I have not seen recorded) fixes beyond 

 further question their family relationship. 



The Ennominae remain an unwieldly mass, but afford no characters 

 which I can detect, that are stable enough to warrant more than generic 

 separation. 



