I9I4-] F. H. GravEIvY : An Account of the Oriental Passalidae. 32g 



Tarquinius than in Pseudepisphenus , i.e. that the peculiarities which have appeared 

 in Pseudepisphenus tend to be accentuated in Tarquinius. Quite in keeping with 

 this is the fact that the scars on the mentum, of which the primary pair has 

 disappeared and the secondary is small in Pseudepisphenus , have completely dis- 

 appeared in Tarquinius. And the outer tubercles of Tarquinius very closely re- 

 semble the right outer tubercle of Pseudepisphenus and that part of the left which 

 is symmetrical with it— all that is required to produce a head shaped like that of 

 Tarquinius from that of Pseudepisphenus , is a shift forwards of the inner tubercles 

 to the anterior margin, where that of the left side would replace the asym- 

 metrical inner portion of the left outer tubercle. The outer tubercles in the I^eptaula- 

 cinae, on the other hand, are simple and more or less acute ; and the lycptaulacinae 

 are the only known Indo-Australian forms to which Tarquinius bears even a 

 superficial resemblance. It is further separated from them by the presence of six 

 well-developed lamellae on each antenna instead of only three, by the structure of its 

 mandibles, and by the absence of scars from the mentum. 



The differences between the Aceraiinae and Gnaphalocneminae seemed very great, 

 so long as ray knowledge of the latter subfamily was practically confined to the 

 information I could glean from previous authors. The Aceraiinae are, undoubtedly, 

 somewhat isolated from a geographical point of view ; and the separation has been a 

 convenient one for the purposes of the present paper. That their recognition as a 

 distinct subfamily on anatomical grounds is less easy to justify than it at first 

 appeared to be, and that it may have to be abandoned, has already been suggested 

 in Appendix I. If Tarquinius is also to be included in the Gnaphalocneminae, it is 

 difficult to see why the Macrolininae and perhaps also the Pleurariinae and even the 

 IvCptaulaciuae should not be included as well. Whether these several series are to rank 

 as subfamilies or as groups, is purely a question of convenience; and it cannot be 

 satisfactorily settled by one who has no special knowledge of African and American 

 forms, 



Pseudepisphenus and Tarquinius appear to have originated from an asym- 

 metrical ancestor within some group of the Gnaphalocneminae as defined at the begin- 

 ning of this paper. There is no reason to suppose that they have had a separate 

 ancestry from symmetrical forms, as is the case with the Aceraiinae. It is, however, 

 impossible to include them in the Gnaphalocneminae without redefining that sub- 

 family. This I could not do without either splitting it up into groups, or else enlarg- 

 ing it to some extent and thus raising the larger issues referred to in the preceding 

 paragraph ; and these I am not in a position to deal with. For the present, therefore, 

 it seems best to regard the genera Pseudepisphenus and Tarquinius as constituting a 

 distinct subfamily Tarquiniinae, distinguished from all others by the strong groove 

 which extends from the upper tooth to the outer angle of both mandibles. 



The genera Pseudepisphenus and Tarquinius may be separated thus: — 

 Inner tubercles situated behind anterior margin of head ; outer 



tubercles asymmetrical .. .. .. .. Pseudepisphenus, Gravely. 



Inner tubercles situated on anterior margin of head as in Lep- 



iaulax; outer tubercles symmetrical .. .. .. Tarquinius, Knwert. 



