116 THE CARABIDAE OF TASMANIA, 



with advantage be restricted to them wlieu describing Carabidae. Umbilicate 

 punctures are often wanting from all the odd irterstices except the ninth.. Dr. 

 G. H. Horn has said that they are wanting from the ninth interstice in the 

 genera Panageus, Micrixys, and Apotomns, but I am not sure that sucli is the 

 case; in Panageua and Micrixys they seem to be hidden by the setosity of the 

 interstice, but I considered I detected umbilicate punctures in Panageus; Apoto- 

 mus shows, in fresh specimens, at least one long sensitive seta near the base 

 and another near the apex of the ninth stria; if these setae are not rubbed off, 

 the punctures from which they rise may be discerned. I regard these sensitive 

 setae of the odd interstices of the elytra as honiologous with the "macrotrichia'' 

 found by Dr. Tillyard on the wings of the Meeoptera; and this gives a leason 

 for their position and taxonomic value. 



Anterior cotyloid cavities. In the first division of the Carabidae. or sub- 

 family Carabinae (here called Carabidae disjunctae) the anterior coxal ca\ities 

 have one opening inwards; in the second division, or subfamily Harpalinae (here 

 called Carabidae conjunctae) there may be either one or two openings inwards; 

 therefore I have further divided the Carabidae conjunctae by this character into 

 Carabidae uniperforatae and Carabidae biperforatae. The division of the in- 

 ward opening of the anterior cotyloid cavities iijto two foramina is caused by 

 .■i chitinous crosspiece which extends at right angles from each .?ide of the cntf^ 

 I'urca . 



Attention may be drawn to an aberrant modification of the biperforate form 

 of the cavities found in the genus Silphomorpla, where the point of each 

 epimerum has moved forward and become attached on each side to the ehitmous 

 crossbar of the cavity; this results in Silphomorpha showing but one opening in- 

 wards, which is not homologous with the single opening of the uniperiorate 

 cavity, but with the anterior foramen of the biperforate cavity. Silphomorpha 

 has the antefurca very short, the posterior part of the anterior coxae more ex- 

 posed than usual, and the posterior opening of the ordinary biperforate cavity 

 completely lost owing to the shifting forward of the epimera. Our othei' Pseu- 

 domoi-phid genus Adelotopus has the ordinary biperforate form of the cotyloid 

 cavities, and the antefurca of usual length. It may be noted, as a case of analo- 

 gous variation, that the anterior coxal cavities in the family Hydrophilidac re- 

 semble those of Silphomorpha. 



The results obtained by the use of the different forms of the anterior coty- 

 loid cavities in the classification of the Carabidae are satisfactory, and a gi-eat 

 help in determining the position and afTinities o^ many genera in the family; 

 but their use causes the arrangement of the tribes to differ greatly from the 

 system now generally recognised. Taking the "Catalogus Coieoptei-orum 

 Europae "(1906) as a standard of the present arrangement of the tribes the 

 plan here adopted brings about the following changes of position. The Scari- 

 tinae, Elaphrinae and Lorocerinae would come first, followed by the Omonhro- 

 ninae and the Carabinae; the position of the tribes from Morioninae to Pori- 

 goninae would remain the same; Granigcrinne, Harpalinae, Zabrinae, Amarinae. 

 Pterostichinae, Masoreinae. and Odacanthinao would follow as members of the 

 Carabidae uniperforatae (but my arrangement of the tribes of this division 

 would not be the same as in the Catalogus). Apotominae. Panageinae, Chlae- 

 iiiinae, Oodinae, Licininae, Lebiinae, Dryptinae, and Brachyninae would be placed 

 in the Carabidae biperforatae. The position of the following tribes in my sys- 

 tem may be indicated :— Anchonderini and Egini to be included in the Otiacan- 



