Male Genital Armature of the Dermaptera. Part II. 525 



rellia st&li, with almost quadrate metaparameres, merely because 

 both have the elytra reduced to small lateral flaps. Again, 

 Euborellia itself, that is, in the old sense, contains two distinct 

 groups, even if we pay attention to the reduction of the elytra only, 

 that is, those like E. stall, E. pallipes, etc., in which the elytra are 

 reduced to mere lateral flaps soldered to the mesonotum, and the 

 other group, E. grecni, etc., in which the elytra are decidedly re- 

 duced, but still meet along the suture, and conceal the whole, or 

 almost the whole, of the mesonotum. In the same way, the 

 apical dilation of the abdomen, the original characteristic of Gono- 

 labis, occurs simultaneously in African, Asiatic, and Australian 

 forms, which are not necessarily closely related, and have totally 

 different parameres. 



But yet, if we were to erect new genera for all the minor 

 groups, arranged according to the shape of the parameres, we 

 should end by having nearly as many genera as species, and it is 

 necessary to choose the middle course. We can begin by dividing 

 them into two main groups. The first has the metaparameres 

 nearly or quite as long as the proparameres, more or less lanceo- 

 late, widest in the middle, and acuminate, three or four times as 

 long as broad. We can at once eliminate Mandex peruviana, of 

 which mention has been made already, with its unusually long and 

 powerful metaparameres ; then we eliminate Anisolabis Fieb., as 

 restricted, with no virga. There remain in the group the Ethiopian 

 Logicolabis Zacher, with long, fine virga, and armoured preputial 

 sack ; the Ethiopian Anisolabis pagana Burr, with unarmed pre- 

 putial sack ; and two Australo-Papuan species, A. littorea White 

 (New Zealand), and the diminutive A. verhoeffi, Zacher (New 

 Guinea), both of which differ from Anisolabis sensu strieto in the 

 presence of the virga. Perhaps each of the last three species will 

 require its own genus ! 



We next come to the second major group, in which the meta- 

 parameres are decidedly shorter than the proparameres, and this 

 group is subdivided in turn into one sub-group, in which the 

 metaparameres are from one-and-a-half to three times as long as 

 broad, and another where they are scarcely longer than broad. 

 Both the divisions are subdivided again and again into a number 

 of small groupings, which at one time I thought might be genera, 

 but I shrink from the responsibility of creating so many new 

 genera, and leave the synoptical table as drawn, and append some 

 cross-groupings, which will, I hope, be of assistance as help-notes 

 for purposes of determination, which is a difficult task, owing to 

 the general similarity of appearance of so many members of the 

 Psalidse. Help-notes, not necessarily in a very scientific form, are 

 often of greater use than carefully drawn dichotomic tables for 

 purposes of identification. 



With the object of making this paper more useful, I have 



