OF THE UNITED STATES. 361 



rita have the antenna nearly equally hairy, while in Morio, the outer joints are much li — 

 pubescent than usual. 



Having eliminated this set, which contains nearly all the anomalous genera of tlitlicult 

 location, our task was reduced to finding sufficient differences to isolate the three remain- 

 ing types, Pterostichi, (including Lebia,) Harpalus, and Chlcenius: alter in ing everj means 

 in my power, I have been compelled to rest the first and third almost upon sexual charac- 

 ters, which are here co-extensive with each group: tin' second group cannot rest upon 

 sexual characters, as we there find such structures variable, oot only in allied genera, but 

 even in the same genus: the only character found to distinguish it, is, that the third joint 

 of the antenme is more or less pubescent, wliile in all the genera of the other two groups, 

 with the exception of Badister and some genera allied to Casnonia, that joint is entirely 

 glabrous. This ditlercnce is not altogether satisfactory, but it seems to be tin best that 

 can be obtained, and if we find the groups into which these divisions maj he separated BO 

 distinct that there will be no possibility of confounding them, we maj use tin-. Blight 

 characters as indicating the existence of affinities in scries, which scries, although manifest 

 in nature, cannot be defined. 



The diagnosis of the groups, as subsequently given, will show that such has been done, 

 and that the groups themselves rest upon characters independent of Bexual modifications. 

 The only cases in which confusion can arise, arc in the osculating instances just men- 

 tioned: where the Lachnophori osculate with Galerita, and Badister with Stenolophus. 

 Between Chlcenius and Pterostichus there is an osculation also, so that for the separation 

 of these groups we must rely on the empirical character that the Chlcenii are all pubescent, 

 and the Pterostichi glabrous. 



Passing now to the third subfamily, Scaritini, we find in it the following types, which 

 are obviously distinct: Ozena, Scarites, Bembidium and Carabus, besidi - Brow us and its 

 allies, and Omophron, the position of winch might be questionable. 



The differences in the parapleural cannot yet be employed, for in the allies ol Scarites 

 we find great variations: the form of the prostcrnum must also be postponed, for Elaphrus 

 and Omophron arc both close relation- of Notiophilus. 



The tibia-, however, present important differences, those of Scant. is being produced V< rj 

 much at the external apical angle, giving rise to the form that is called palmated: in tin- 

 group, and in one or two genera (Prouiccocrnathus) which, though not belonging to the 

 group, show very great affinities with it, the para pleurae are unusual in form, bo that 

 throwing out for the present Promecognathus, we can get good definitions for the lour 



principal types, O/.a-na, Scarites, Heuhidium, and Carabus. 



With regard to Omophron, on examination, there is found to be no important diffe- 

 rence between it and Elaphrus, excepl in the form of the sternum; and as we alreadj 

 know the variations in that part of the !>odv by comparing Elaphrus and Notiophilus, it is 

 safer to place Omophron with the Carabi, to which its simple anterior tibia and para- 

 pleura show it- affinity. With regard to the remaining gen< ra not included in the lour 

 groups, they are Promecognathus, Metritis, I»ro-cu-, and I laplo< lnle, and BOme foreign 

 genera, considered as osculant between Mono and < huena, \\ ith both ol w hich ind< i d thej 

 ^ ol. \. 6 i 



