CALIFORNIA ACADEMY OP SCIENCES. 335 



general terms may be stated to be an aggregate of species 

 possessing in common a character, or a certain assemblage 

 of characters, considered by its author to be of sufficient 

 stability and persistence to distinguish it as an isolated 

 group. I say "considered by its author," because it is this 

 individuality in the opinion of specialists on the one hand, 

 and our imperfect knowledge of nature on the other, which 

 prevent the assignment of a definite weight or value to the 

 characters which have been adopted in the separation of 

 genera as they at present exist; these in many cases have 

 been founded upon comparatively trivial characters, and 

 more or less on the score of convenience. In fact when the 

 Coleoptera have been exhaustively collected, it will proba- 

 bly be found that all genera are more or less arbitrary divis- 

 ions, as species must in many cases be discovered with 

 intermediate characters, of whatever nature these may be, 

 showing a gradual progression from one to another. In 

 short, that there is no such thing in nature as a rigorously 

 limited aggregate of species, is, I believe, a widely accepted 

 opinion; therefore all genera must be more or less artificial 

 and instituted primarily in order to secure a natural and 

 systematic arrangement and succession of the species, and 

 incidentally to enable these to be easily identified. 



If this be granted there can, in the opinion of the writer, 

 be no valid reason for the rejection of the genus Hemistenus 

 Mots. (Areus Cas.). This is surely an instance where a divis- 

 ion on the score of convenience is greatly to be desired, and 

 is at the same time fairly warrantable from structural con- 

 siderations. That there are a few forms which are interme- 

 diate and as it were connecting links between the genera 

 Stenus and Hemistenus, is, as above indicated, no more than 

 must be expected, and even with these intermediate forms 

 (which, however, are not very evident in the American 

 fauna), the two genera are apparently much more definite 

 than a multitude of those which already exist, and which 

 are considered well established, especially many of the 

 Harpalinide genera. 



