BY THOMAS G, SLOANE. 455 



buccal fissure equally wide with similar posterior angles, and I 

 should expect the same form to occur in P. distinctus, SI., the 

 only other nearly allied species. It should be noted that I have 

 seen no undoubtedly male specimen belonging to the small group 

 which these three species constitute, for the type specimen of 

 P. ambiguus proves to be the 9> and not the ^ as I thought, and 

 I did not determine the sex of the type specimens of P. distinctus. 



Note. — In the type specimen of P. ambiguus there is a strongly 

 impressed round fovea on the apical dilatation of the inflexed 

 margin of the pronotum; this does not occur in P. angidiceps. 



Tribe HARPALINI. 



The classification of the Harpalini is perhaps the most difficult 

 to elucidate of all the tribes among the Carabid£e,*and it appeai^s 

 to me that the complexity of the subject has been increased by 

 systematists having placed so much reliance for classificatory 

 purposes on the secondary sexual characters of the male, particu- 

 larly the vesture beneath the joints of the tarsi. Although the 

 vesture of the under side of the tarsi is useful in helping to 

 determine the affinities of species, and often of genera, being 

 usually similar throughout each tribe, yet, in the classification of 

 the Harpalini its use has been pushed to lengths that have added 

 to the complication of the subject rather than diminished its 

 intricacy; so much is this the case that it seems impossible, under 

 the existing system, to refer a Harpalid of which the male is 

 unknown with any certainty to its proper genus, or even to 

 determine the group of genera into which jt should fall. Any 

 classification which is founded on secondary sexual characters 

 seems to me to be so artificial that an attempt to replace it is 

 always worth trying; therefore I offer the following table (though 

 aware it is far from perfect) in the hope that by its aid the limits 

 of the genera among the Australian Harpalini may be more 

 accuratel}' defined than seems to be the case at pi-esent. 



* Vide Dr. G. H. Horn's remarks, Trans. Am. Ent. Soc. 1881, ix. p. 175. 



