80 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 5° 



ogy of the costal vein of the hind wing and the arrangement of the 

 distal hooks, gives in the main the following arrangement : 



I. — Costal vein divided at the base. 



A. — Upper branch of the costal vein marginal, reaching to the middle of the 

 wing, where it again unites with the lower one. 



Subdivisions are distinguished according to the following principles : 

 Cross-vein branches from the costal vein at or behind the point of 

 junction of the two branches of the costal vein. 



Series of distal hooks beginning before, at, or behind this point of 

 junction. 

 B. — Upper branch of the costal vein marginal, after a short space becoming 

 very thin or disappearing altogether. The lower branch reaches the 

 wing-margin behind the middle. 

 Subdivisions according to the following principles : 

 End of the upper branch of the costal vein distinct or indistinct. 

 Wing-membrane outside the costal vein in the region of the distal 

 hooks visible or invisible. 

 11. — Costal vein not divided at the base ; simple ; behind the middle of the 

 wing not marginal. 

 Subdivisions according to the following principles : 

 Cross-vein branching from the costal vein behind the middle; 

 Series of distal hooks begins before or at this branch. 

 777. — Costal vein not divided at the base ; marginal ; unites behind the middle 



of the wing with the II longitudinal vein. 

 A. — The two veins, after their junction, continue as one. 

 Series of distal hooks begins before this junction. 

 Subdivisions according to the following principles : 

 Series of hooks single or double. 

 B. — Costal and II longitudinal vein forming a loop by again dividing after 

 their junction and thereupon again uniting. 

 Series of hooks always double. 



To ascertain the family to which an undetermined Hymenopteron 

 belongs, the table of Staveley is not suitable. The reasons for this 

 are the following : Only a few families are restricted to one division 

 (Vespidse, I. A; Ichneumonidae, I. B ; Apidae, Formicidae, II) ; many 

 families and even subfamilies are distributed over several divisions ; 

 so the Sphegidse over I. A and I. B, the Tenthredinidae over II., III. 

 A, and III. B, etc. Furthermore, in some of the divisions represent- 

 atives of the most heterogeneous families are united, e. g., II. Apidae, 

 Crabronidae, and Tenthredinidae. Finally the number of species indi- 

 cated as examined by Staveley is exceedingly small. Cynipidae and 

 Braconidae are not mentioned by her at all. Moreover, no necessity 

 exists for the use of the differences in the arrangement of the hooks, 

 spines, etc., other much more conveniently examined characters, as 

 the wing-venation, the structure of the legs and antennae, etc., being 



