46 MEMOIRS ON THE COLEOPTERA 



the most natural that can be devised, relate to the existence of 

 one or two supra-orbital setse extremely significant in the second 

 subdivision of the family but of no value in the first and the 

 presence or absence of a postero-external elytral plica. These 

 very important discoveries give us the means of resolving the 

 family into several very clearly demarcated sections. I would 

 suggest in this connection that the genus Pseudomorpha be separated 

 from the Carabidae to form a distinct family. It does not fall in 

 line with the true Carabids very well either in structure or facies. 



The Harpalinse, as here considered, embrace all Carabids in 

 which the mes-epimera fail to attain the coxae and are narrow and 

 parallel in form and also in which the head has but a single supra- 

 orbital seta, the mandibles devoid of an external setigerous punc- 

 ture, the posterior coxas contiguous and the elytra without a postero- 

 external plica. I have no means of verifying the opinion of Dr. 

 Horn (Tr. Am. Ent. Soc., 1881, p. 175) that the singular African 

 genus Glyptus, having no seta on the second labio-palpal joint, 

 is a component of the subfamily, but am inclined to believe that the 

 rernarkable group of genera clustering about the European Ditomus 

 should be excluded and form a subfamily by themselves, because 

 of the strongly pedunculated body, long antennas, occasional very 

 striking modifications of the epistoma and mandibles, the peculiar 

 coarse sculpture of the body and the marked departure in general 

 habitus. Although most of the Ditomid genera are represented in 

 the material at hand, I have therefore thought best not to include 

 them within the scope of the subfamily as here considered. All 

 of the European tribes of the subfamily, as thus restricted, occur 

 in North America, but there are a few American tribes such as 

 Cratocarini which do not occur in the old world. 



In the table of tribes given below, it will be noted that the 

 division heretofore proposed into three groups of genera, depending 

 upon the structure of the male tarsi, has been abandoned and a 

 succession of tribes defined upon more restricted sexual characters 

 of the same kind, the two principal divisions, however, being based 

 upon the setee of the second labio-palpal joint, first suggested by 

 Bates. The full importance of this character, which widely shifts 

 the positions of several important genera such as Polpochila and 

 Agonoderus into more congenial surroundings, escaped the attention 



