314 



males and 64 females. Separating these according to the number of 

 antennal joints (a character upon whichMr.Ashmead lays great stress) 

 gave the following results : 



Males: 



14 antennal joints, 18 specimens. 



15 antennaljoints, 37 specimens. 



16 antennal joints, 1 specimen. 



15 joints in one antenna and 16 in the other, 1 specimen. 

 Females : 



12 antennaljoints, 7 specimens. 



13 antennal joints, 54 specimens. 



14 autennal joints, 1 specimen. 



12 joints in one antenna and 13 in the other, 2 specimens. 



From this it will be seen that the normal number of antennal joints 

 in these specimens is fifteen for the male and thirteen for the female, in 

 each sex the number varying by one joint more or one less than the 

 usual number, the tendency being in the direction of a less number of 

 joints. In those having an unequal number of joints in the antennae 

 of the same individual this is evidently the result of a consolidation of 

 the last two joints in one of the antennse, since the last joint in the an- 

 tenna containing the fewest joints is always longer than the last joint 

 in the opposite antenna. This process, however, will not account for 

 the varying number of antennaljoints in the different individuals of the 

 same sex; thus, in the male with sixteen antennaljoints the last joint 

 is comparatively as long as it is in the males with only fourteen antennal 

 joints. As a rule, the specimens with the fewest antennal joints are 

 smaller than the others, and this is also the case with the specimens 

 examined by Mr. Ashmead. 



Taking the 37 males with 15-jointed antennae above referred to, I am 

 unable to discover any character or characters by which they can be 

 separated into distinct species, or even into varieties. The last anten- 

 nal joint varies all the way from slightly shorter to one-half longer than 

 the preceding joint ; petiole of abdomen usually yellowish, sometimes 

 with a dusky spot above posteriorly, this spot becoming larger in dif- 

 ferent specimens until the entire upper side of the petiole is blackish, 

 and this color is rarely still further extended until the entire petiole, 

 except a small spot on either side, is blackish ; middle coxae usually 

 blackish, except the apex, but this color varies in extent in the differ- 

 ent individuals until only the extreme base of the coxae is black ; hind 

 tarsi vary all the way from as long as to much longer than their tibiae; 

 the second section of the radius varies from one-half as long to nearly 

 as long as the first section ; the transverse cubital nervure is sometimes 

 hyaline in the middle or is wholly hyaline, and in one specimen it is 

 entirely wanting in one wing, while in the opposite wing only a frag- 

 ment of it remains, but in another specimen it is entirely absent in each 

 wing. 



The 54 females with 13-jointed antennae above referred to vary to the 



