Inn Annuls Entomological Society of America [Vol. Ill, 



The jjurtiun of the head, showing behind the eompound eye. 

 is termed the cheek, in this paper. The width of the cheek 

 does not vary a great deal. The central portion, however, 

 (that portion in the region of the attachment of the head to the 

 thorax) is slightly wider than other parts. 



LAHRIM. 



The labrum is attached to the lower, inner edge of the cly- 

 peus, leaving the outer edge of the clypeus well defined. In 

 preserved specimens, the labrum is usually bent backward, at 

 nearly right angles to the clypeus, with the mandibles closed 

 over it, so that it is not accessible for study. For this reason, 

 it has not seemed advisable to use it for analytical purposes. 



MOUTH TARTS. 



None of the mouth parts seem to' be useful for the determina- 

 tion of the species and their description therefore is not included 

 here. I might say. however, that possibly the mandibles, 

 labial palpi, and maxillary palpi w^ould be of use in the determi- 

 nation as in most of the insects examined, the palpi show to the 

 full extent. Maxillary palpi five jointed, labial palpi four 

 jointed. 



The mandibles are somewhat curved, having two teeth, 

 which are blackish to their bases. When the jaws are closed 

 they overlap. 



ANTENNA. 



The antennae are long. The basal portion of the bulb of the 

 I)roximal segment or scape is rather small and articulates with 

 the head in a socket. Its diameter, at this end. is about equal 

 to its length to where it unites with the enlarged portion of the 

 scape, but it is somewhat narrower toward the place where it 

 enlarges into the scape proper. The bulb appears to be an 

 entire segment, but this is not the generally accepted view. 

 The scape which is the stoutest segment of the antennae is some- 

 what smaller at its base, gradually increasing to the base of the 

 next segment, the pedicel. This seems to extend down into the 

 scape for some little distance. There is also a third segment 

 before we come to the filament ])ro])cr which is nearly twice as 

 l)road as long, and is called the ring joint. This seems to be 

 rather unusual. The filament which consists of about fifty- 

 four segments has those near the base a little longer than the 

 others, gradually decreasing in length as we go towards its 

 extremity. 



