ICHNEUMON-FLIES — GELINAE : MESOSTENINI 11 



subtribe. The siibtribe Echthrina nevertheless seems to be a de- 

 finable and useful grouping. 



A number of genera are occasionally or habitually parasitic in the 

 nests of wasps or bees. Some, like Aritranis and Xylophrurus, are 

 normally borer parasites, and the parasites of Aculeata among them 

 have made the sHght shift from ovipositing into borers in twigs to 

 ovipositing into wasp or bee nests in twigs. Such a shift seems to be 

 easy, and to have such little significance that some ichneumonid 

 species may parasitize indifferently either borers in twigs or aculeate 

 nests in twigs. Others have made a more profound biological shift 

 and usually the host preferences of entire genera are involved. These 

 concern cases where the external characteristics of the nest are different 

 from the usual hosts of the tribe, and the ichneumonids have developed 

 special characteristics that fit them to locate such hosts, oviposit 

 through the special texture of the nest into the host itself, and as a 

 newly matured adult to cut a way out through the special nest texture 

 to freedom. Some of the genera involved, like Toechorychus and 

 Pachysomoides , are only a little modified for such host relations. In 

 others the changes have been more profound, resulting in characters 

 that set them apart from other mesostenines. The more modified 

 genera may be divided into two groups, which are here recognized as 

 subtribes: the Sphecophagina adapted to parasitizing the paper nest 

 cells of social Vespidae, and the Nematopodiina adapted to parasitiz- 

 ing the mud cells of sohtary Vespidae, Psammocharidae, and 

 Sphecidae. In the Sphecophagina, the short ovipositor and short 

 mandibles are obviously adapted to its peculiar hosts. In the Ne- 

 matopodiina, the longer ovipositor, usually with coarse apical teeth, 

 and the picklike mandible are adapted to drilling through a mud-nest 

 wall for ovipositing and to cutting out as an adult. Judging from the 

 diversity of the ovipositor of the nematopodiine genera, however, the 

 host range may include some things other than aculeate nests. (The 

 ovipositor of Nemafopodius itself, for example, is quite different from 

 that of the rest of the tribe. There are no rearing records for this 

 genus.) 



Thus we classify certain host-specialized gi'oups of mesostenine 

 genera as distinct subtribes, leaving the great bulk of the genera, 

 including probably the less specialized representatives of all or most 

 of the evolutionary lines, in the large subtribe Mesostenina. Since the 

 main characters of the new subtribes are clearly adaptive, it is reason- 

 able to suspect, and some evidence supports the supposition, that not 

 all of the new subtribes are monophyletic. Convergence may have 

 obscured some of the true relations. The present subtribal divisions 

 should prove useful, however, and are steps toward a natural classifi- 

 cation. 



