AMERICAN COLEOPTERA. 91 



On the genera of CARABIDvE with special reference to the 

 fauna of Boreal America. 



BY GEORGE H. HORN, M. D. 



Before entering the subject of the present essay it will be useful to 

 consider some preliminary matters about which there is still some dis- 

 cussion and differences of opinion. 



The Carabidje forms one of the members of the Adephagous series of 

 coleoptera which is readily recognized by the predaceous character of its 

 mouth parts, its slender antennae (except in Gyrinidae), pentamerous tarsi 

 and the structure of the first abdominal segment which is in all cases 

 divided or hidden by the posterior coxae in such a manner, that it is 

 entirely lateral, rarely appearing as a small triangular piece between the 

 posterior coxae. 



The ventral character is an important one and it prevails without 

 exception in the entire series. If we examine the Gyrinidae, however, 

 which the books all agree in saying have six ventral segments, the first 

 segment will be found apparently very long and passing entirely across 

 the abdomen without interruption by the coxae. A more careful ex- 

 amination will show that this segment is really composed of two closely 

 united, but with traces of the suture visible near the coxae as I have 

 illustrated on PI. Ill, fig. 9. The Gyrinidae therefore have seven ven- 

 tral segments. 



With an apparent exception thus disposed of the limits of the ftimilies 

 of the Adephaga are to be considered and here is the point where the 

 greatest diversity of opinion prevails, especially with reference to the 

 Haliplidae and the two aberrant genera Amphizoa and Fe/obim. 



The Haliplidje are placed by Lacordaire in the Dytiscidae rather under 

 protest, while subsequent authors have removed them in a more or less 

 decided manner either as a separate family or sub-family. In a pre- 

 liminary sketch of a new classification of Dytiscidae, ( Comptes-rendus 

 de la Soc. Ent. Belg. Sept. 4, 1880), Dr. Sharp says: "I exclude the 

 Haliplides from the family and leave for the Carabophiles to decide 

 whether they should be considered Carabidae or form a distinct family." 

 They should form a separate family ; my reasons will be given hereafter. 

 Amphizoa immediately concerns our fauna. It was originally de- 

 scribed as typical of a distinct family by Dr. LeConte, (Proc. Acad. 

 1853, p. 227 — 8), and notwithstanding the opposition of Schaum the 



