4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 94 



one set of characters, the only possible exception to this being the 

 work of Eichelbaum and of Sharp and Muir on the genitalia. 



Material 



The material available for this study includes chiefly specimens 

 from North America supplemented by additional ones from Central 

 America and Europe. Wherever possible one of the common species 

 was selected. 



Eichelbaum (1909) hsted 15 subfamilies of the Staphylinidae. In 

 these he recognized 43 tribal groups. The present study includes 

 representatives of 10 subfamilies and 25 tribes. The subfamilies of 

 which representatives were not available are : Megalopinae, Lepto- 

 typhlinae, Euaesthetinae, Trichophyinae, and Cephaloplectinae. Be- 

 sides these, the tribes not represented include : Neophonini, Apa- 

 teticini, Platyprosopini, Pygostenini, Trichopseniini, Deinopsini, Gym- 

 nusini, Myllaenini, Pronomaeini, Diglottini, Hygronomini, and 

 Digrammini. 



The species selected in each tribe are not considered as actual rep- 

 resentatives or types but merely as members of that group and ex- 

 amples of it. In several cases more than one species has been used 

 within a tribe because of the great variation in certain characters, 

 but lack of time and space has prevented the further extension of 

 the list. 



It is very much to be regretted that members of the subfamily 

 Aleocharinae are so difficult of identification while so abundant every- 

 where. Only the larger tribes are represented in this material, and 

 this means that the aberrant and unusual forms have not been studied 

 or included in the comparisons. 



All the American species were identified by the writer. Most of 

 the European species were obtained from Emmerich Reitter of Vienna. 



preparation of material 



The preparation of material so that it can be adequately studied 

 is one of the most important steps in the solution of any systematic 

 problem. In the treatment of the specimens first consideration should 

 be given to the making available of all characters which might be 

 of value in the study. How far this ideal is from the condition ac- 

 tually attained in much systematic work has been discussed at length 

 by Ferris (1928), and a similar inadequacy is often shown in morpho- 

 logical studies. 



