80 r^prii. 



group. This nerve has simply disappeared in most of the forms of that p-roiip 

 to which s/uir/n and Jilicis stand nearest, and it does not seem impossible ihat 

 the ancestors of these two forms should also have lost it, and that subsequently 

 a new " adventitious " nerve should have been developed running- in a different 

 direction. At any rate, these two types of neuration in the lanceolate cell 

 differ so widely that they oui,'-ht not to be tabulated as though the}' were 

 identical. 



5. Cameron describes the claws of s/(rtr/»' as having a " subapical tooth." 

 Here, again, i cannot follow him. Tiiere is a sort of triangular dilatation of 

 tlie inner margiu near its centre (see fig-.) which can hardly be seen without 

 pushing aside the " j ulvillus."' But it is scarcely to be called a "tooth," and 

 its situation is certainly not "subapical." 



0. Of the antennae Cameron says merely that they are of the usual form. 

 It should h.ive been added that the basal joint is much longer than the second, 

 which is very transverse. The third and fouith joints are subequal, the third 

 perhaps a very little longer than the fourth. 



The face is short and broad. Viewed from in front its npper or posterior 

 horizon rises, lifting the ocellar area, considerably above the tops of the eyes; 

 the cheeks (between the eyes and the mandibles) are pretty long and evidently 

 CDUvergeut. The mandibles are almost exactly as in cinyulata — much dilated 

 berjeath and rounded off to their apices, which are bitid. The saw-sheath is 

 small and inconspicuous, snbtriaugular in the lateral view. Seen from above 

 it is narrow throughout, praciic.iUy linear ! 



Although sharjji is clearly a Sfroin/yloffasfer in the earlier and 

 wider sense of that term, it can hardly be placed in that genns as 

 limited by present systematists. The other British species formei'ly 

 included in Strongjjlngaster are now distributed between four sup- 

 posedly and, perhaps, really distinct genera — namely, Strongylogasier 

 Dahlbom (cingiilafn, xanthocera) ; Pseudotaxonns A. Costa, 1S9-1 

 {Jtlicis) ; Stromhoceros Konow {clelicatulus) ; and Thrinax Konow 

 {macnhi, mixtd, configini). Jn a limited fauna like our own so much 

 subdivision is rather inconvenient, but when exotic forms have to be 

 dealt with it can hardly be avoided. According to the above arrange- 

 ment, sliarpi cannot well be placed anywhere but in Pseudotaxonits, 

 unless, indeed, yet another new genus is to be erected for its reception, 

 which I am far from wishing to propose. Yet I cannot but think that 

 it is rather " unequally \'oked " with JiJicis. The " habit " of the two 

 insects is wlioUy different, and though the venation of the fore wing is 

 identical in and peculiar to both, that of the hind wing is exceedingly 

 different. Also a part of the present definition of Pseudotaxomis is 

 " Pentagonal area indistinct," whereas — as I have said alread}- — it 

 appears to me to be " distinct " in sliarpi. Apart from venation 

 characters, it is clearly separable from Stromhoceros by the structure 

 of its antennae, and from Tlir'niax and also from Sfroiigiflogdnfer by 



