490 U. S. NATIONAL MUSEUM BULLETIN 216 



rather short to elongate and slender; second tergite nearly always 

 with an oblique basal groove on each side cutting off basolateral 

 corners, and often with other grooves or impressions; second and 

 third tergites variously sculptured; apical part of ovipositor cylindric 

 or slightly depressed, the lower valve with about eight ridges, basad 

 of which there is a roughened area (fig. 331,h). 



The species of Xorides are many, and are distributed over almost 

 the entire world. They form a natural, strongly marked group that 

 is easy to recognize. Among them, however, is great structural 

 diversity and the species may be divided into a number of groups of 

 varying size, distinctness, and ease of definition. Most of the more 

 distinct groups have had generic names proposed for them, and 

 whether and which of these to recognize as separate genera has been 

 variously decided by various authors. We prefer to include all of 

 them under a broadly defined genus Xorides. It seems desirable, 

 however, to recognize some of the more distinct groups as subgenera. 

 In the first of the keys below is indicated our version of the sub- 

 genera of the World. In the Nearctic region, there are only two 

 subgenera, Exomus and Xorides, so the subgeneric arrangement has 

 relatively little application to our fauna. It is very useful, however, 

 for dealing with the species of the Old World tropics. 



We are indebted to Mr. J. F. Perkins of the British Museum 

 (Natural History) for permitting us to study his manuscript key to 

 the subgroups of these insects. Although we have verified his work 

 and modified it, most of the better characters given below were sug- 

 gested to us by his key. The subgeneric synonymy is according to 

 our studies of the types of the genotypes, of which we have seen all 

 but the types of Ichneumon indicatorius Latreille, Neoxylonomus an- 

 tennalis Szepligeti, and Xylonomus gracilicomis Gravenhorst. The 

 type of indicatorius is lost, or destroyed. The type of gracilicomis is 

 destroyed. It was lent to Clement along with other Gravenhorst 

 types of this group for his revision of the European species. These 

 types were returned by Clement to the Gravenhorst collection in 

 Wroclaw, but in the case of gracilicomis, only the pin with labels 

 was present in 1958. The specimen itself was not on the pin. 



We are not entirely satisfied with the subgeneric arrangement 

 below, but present it as a basis for future studies, and to bring some 

 order out of the general confusion. The species of Xorides tend to 

 be so scarce in collections that it will be some time before they will 

 be well enough collected to permit a "definitive" classification. The 

 main problem is to predict whether the groups that are definable on 

 the basis of the few species known will remain definable when a 

 larger portion of the fauna is available for stud} 7 , or when the known 

 species can be gone over in more detail. So many groups that ap- 



