NO. 2334. NORTH AMERICAN ICHNEUMON-FLIES— CUSHMAN. 269 



KEY TO NORTH AMERICAN GENERA OF CEEMAaTINI. 



1. nind femur with a tooth beneath 2. 



Hind femur normal 3. 



2. Stigma narrow lanceolate, radius at or before middle; petiole with tergite entirely- 



enclosing sternite except at extreme apex and base, its lateral edges fused be- 

 neath, the suture obsolete; tooth of hind femur at apical fourth or fifth; com- 

 pression of abdomen including second tergite; very long, slender species 



Eiphosoma Cresson. 



Stigma broad triangular, radius more or less behind middle; first tergite not en- 

 closing sternite; tooth of hind femur at or before apical third; second tergite 

 not compressed; short, stout species Pristomerus Curtis. 



3. Areolet present; upper surface of abdomen in female very largely covered by three 



segments, those beyond third very short Oligotmema, new genus. 



Areolet absent; segments beyond third not especially ret acted 4. 



4. (Key to this group already published.) 



Genus EIPHOSOMA Cresson. 



Eiphosoma Cresson, Proc. Ent. Soc. Phila., vol. 4, 1865, p. 52, fig. 5. 

 Xiphosoma (Cresson) Szepligeti, Gen. Ins., fasc. 34, 1905, p. 4. 



As stated above Cresson considered this genus to be related to 

 Cremastus. Because of its toothed femora Ashmead ^ placed it in 

 his tribe Pristomerini; but later,^ apparently because of its narrow 

 stigma, he removed it to the Anomalini, where it is obviously dis- 

 cordant. Morley^ follows in the footsteps of Ashmead, though 

 calling attention to the obvious relationship to Pristomerus. Szep- 

 ligeti* erected his subfamily Xiphosominae for this genus and his 

 Xiphosomella, but maintained it in close proximity to the Anoma- 

 linae, though calling attention to it under the Pristomerinae. Ex- 

 cept Cresson and Roman apparently no one has thought of allying 

 it with Cremastus, all writers ascribing too much importance to the 

 shape of the stigma and the femoral tooth, neither character in the 

 opinion of the writer of more than generic value. Schmiedeknecht," 

 however, while si-ill using the form of the stigma as his prime char- 

 acter, in his arrangement places the Xiphosomini as Tribe No. 13, 

 the Pristomerini as No. 9, and the Cremastini as No. 10, the Pori- 

 zonini and Plectiscini falling between the Cremastini and Xiphos- 

 omini. 



Viereck divided the genus, making Eiphosoma pyralidis Ashmead 

 the type of his genus BracJiyxipliosoma. His description consists of 

 the statement of two characters by which it is said to differ from 

 Eiphosoma. Both of these characters, the length of the malar space 

 and the distance to which the propodeum overlies the hind coxae, 

 are very variable and are not always associated in the combination 



I Trans. Amer. Ent. Soc, vol. 23, 1S96, p. 191. 



» Proc. U. S. Nat. Mus., vol. 23, 1900, p. 88. 



3 Rev. Iclin., pt. 2, 1913, p. 51. 



* Gen. Ins., fasc. 34, 1905, p. 2. 



<* Opusc. Ichn., fasc. 18, 1908, p. 1409. 



