1. ST RATI O MY I D^ 59 



invariably connects at or near the wingmargin witli the anal vein and thereby 

 causes a closed anal cell, and in the Fachi/gastrimv and Berhiie it curves down 

 sharply and connects v^dth the anal vein far before the wingmargin. The absence 

 of an ambient vein, by which term is meant a thin vein continued after the end of 

 the lower branch of the culiital fork all round the wingmargin to tlie alula is a 

 peculiarity in the venation of the Stratiomi/idce ; in the Berina^ this vein is present 

 for a short distance, and in the Xylomjiince it extends to the end of the second 

 veinlet from the discal cell, but in the allied Xylophagiiue it is complete though 

 faint. The crowding of the strong veins together near the costa with a correspond- 

 ing faintness of the posterior veins is characteristic of the more typical subfamilies, 

 and altogether the venation of the Stratiomi/uke (with the exception of the 

 Xi/h7ni/in(e) should enable them to be at once distinguished from any other family 

 of the Diptera. 



*»^ I am unable to agree with Girscliner's study of the venation of the Nutacantha, Ann. Mus. Hung. iv. 

 279, V. (1906), in which he is of opinion that the vein which 1 call the upper branch of the cubital fork is in 

 reality a lower branch of a radial fork, and that the radial vein has been anastomosed with the cubital 

 along what 1 call the stem of the cubital fork ; Girschner has arrived at this opinion through a study of the 

 convex and concave veins, but I think he has overestimated their value, as his fig. 5 (^.iurgus) would make the 

 prEefurca a basal portion of the cubital vein, while in all his other ligures it would be the basal portion of the 

 radial vein. Also I cannot agree with him that the ordinary radial vein is missing in Ncmutelus : 1 am of 

 opinion that it simply runs along very close to the subcostal vein. 



Sho7't Tahle of British Subfamilies of Stkatiomyid^?!;. 



1 (8) Abdomen with only five or six obvious segments. 



2 (5) Small cross-vein absent. 



3 (4) Discal cell emitting only two veinlets (fig. 77) irrespective of the 



upper branch of the postical vein. PACHYGASTRiNiE. 



4 (3) Discal cell emitting three veinlets (fig. 78) irrespective of the 



upper branch of the postical vein. ClitellariN/E. 



5 (2) Small cross-vein present. 



6 (7) Antennae without any terminal style or arista, or at the utmost 



with a short one. SiRATiOMYiNiE. 



7 (6) Antennae with a long thin terminal arista. Discal cross-vein 



connecting the discal cell with the prcefurca. Sarginv-e. 



8 (1) Abdomen with at least seven obvious segments. 



9 (10) Praifurca commencing opposite the base of the discal cell; fourth 



posterior cell wide open. Berin^e. 



10 (9) I'raefurca commencing far before the base of the discal cell; fourth 



posterior cell closed. Xylomyin.e. 



The CUtellariiue and Stratiomi/inie are perhaps unnaturally separated, as the 

 presence or absence of the small cross-vein is sometimes a debatable point. 



Table of British Subfamilies of Stratiomyid^. 



1 (8) Abdomen with only five or six obvious segments (figs. 85, 87, etc). 



Cubital vein often not forked or with only a small indistinct 

 very short fork (fig. 83). Tibise never spurred. 



Species (except in some Sargince) with a broad short ovate 

 abdomen. Flagellum with not more than six annulations. 



2 (5) Small cross-vein absent (fig. 83), because the upper branch of the 



postical vein forms a portion of the lower margin of the 

 discal coll. 



