OF WASHINGTON. 57 



Basal joint of hind tarsi twice as long as the four follow 

 ing joints united; the first branch of radius shorter 



than the second Eurypterna Fonst. 



(Type Pachyl. cremieri Breb.) 



Basal joint of hind tarsi scarcely one-third longer than 

 the four following joints united; the first branch of 



radius much shorter than the second Pachylomma Breb. 



(Type P. buccata Bre"b.) 



Notwithstaning the concensus of opinion in favor of the group 

 belonging to the Braconidce by such authorities as Nees, West- 

 wood, Wesmael, and Forster, the latest writer on this family, the 

 Rev. T. A. Marshall, in his " Monograph of the British Bra- 

 conidas," 1885, evidently holds' different views and makes no 

 reference to the group, except in the following curt sentence : 

 ''Excluding Forster's Pachylommatoidce, as belonging to the 

 Evaniidce, we will treat the Braconidce under six divisions, each 

 including a number of subfamilies." He then follows with his 

 tables, and has accepted, as subfamilies, all of Forster's families, 

 except the Pachylommatoidce . 



During my visit to Berlin, in the winter of 1 889-^0 , I saw for 

 the first time, in the Royal Berlin Museum, a specimen of Pachy 

 lomma buccata, and at once recognized its relationship to a Bra- 

 conid in the U. S. National Museum, which I had only a short 

 time previously described as Wcsmaelia rileyi, in the subfamily 

 Euphorince. 



At the time of my drawing up the description I did not feel 

 entirely satisfied that it was properly placed, and remarked as fol 

 lows : u This remarkable insect agrees with the definition of this 

 genus, but seems out of place in the group, and appears more 

 closely allied with the group Aphidiince, where it may ultimately 

 be placed." 



That it was anything else than a Bracon, however, never once 

 entered my mind, and now, after four years' study in the family, I 

 am convinced the Pachylommatoidcc are Braconids, and should 

 be considered as a subfamily allied to the Euphorince and the 

 Aphidiince and of equal rank. 



The mere fact that the abdomen is attached to the metanotum 

 some distance above the insertion of the hind coxae is in itself 

 not sufficient, in my opinion, to separate the group from other 

 Braconids, especially as a few other Braconids show this charac 

 ter to a still greater degree. For example, the genus Cenocoelius 

 Haliday (which equals Aulacodes Cress. = Laccophrys Forster 

 = Capitonius Brulle = Promachus Marshall) has the abdomen 

 attached far up on the metathorax as in the genuine Evaniids. 



I propose, therefore, to accept Forster's family Pachylomma- 

 toidce as a subfamily in the family Braconidce, and as our species, 

 so far discovered, do not fit exactly into the two European genera, 



