AMERICAN LKPIDOPTERA. 247 



(Tiipulu) the anteniue are doubly bi})ectinate, having two pectina- 

 tions at tlie summit and two at the base of each segment ; this is 

 also the case with the pectinate species of Erannis, a fact not here- 

 tofore noticed. In many species the pectinations are at the summit 

 of the segments, and in others at the base only. This tendency is 

 shown in the species which do not have bipectinate antennae, as well 

 as in the females, the tend('n(;y of the ciliations being to gather into 

 tufts at both suiiiiuit and i)ase as in K('tropl><, etc., or at bottom, or 

 at top oidy. 1 have not been able to make any exclusive generali- 

 zations as y(;t, but the Sterrhinie seem thus to separate from the 

 Geomctriiuc, the former having the pectinations at the bottom, the 

 latter at the top of the segments. In the Ennominie the species 

 having the pectinations at the top of the segments generally corre- 

 spond with those which have the fovea at the base of the fore wings 

 in the male, but this is not always the case. A few species have 

 the antennie serrate, and a few others have them flattened and more 

 or less lamellate ; generally with this structure the antennjc are naked, 

 at least below. I have found only one instance of unipectinate an- 

 tennse among our species, and this is the 9 of Gonodontis hijpo- 

 chraria H.-Sch., but I am not sure that any emphasis can be placed 

 upon this fact. In most species the pectinations on one side are 

 shorter than on the other, and in this species in the female the longer 

 pectinations have apparently just begun, or are just losing their 

 existence, while the shorter ones have not begun, or have ceased to be. 



There are also very considerable differences in the position and 

 shape of the tibial epiphysis. In many species it is below the middle 

 of the tibia and short. In others it is above and often reaches con- 

 siderably beyond the end of the tibia ; sometimes it is considerably 

 tufted. 



The tarsi vary considerably compared with the tibite and witli 

 each other. They are often spinulated, sometimes strongly so. 



These characteristics, so far as my observations go, are compara- 

 tive only. I have as yet not been able to make use of them in clas- 

 sification, but suggest their variability that the future student may 

 make use of them as a larger and better study of material may 

 warrant. 



I have been able to make nothing of the ocelli ; other students 

 have seen them in a few species, but I have to confess I have never 

 yet seen the organ developed in a single Geometer, though for the 

 purpose of study I have bleached hundreds of heads, and in some 



TRANS. AM. ENT. SOC. XXIII. JULY, 1896. 



