06 [Arril, 



writer, the prosternal plate between the front coxae also bears a few 

 short bi'istles or hairs on each side margin (Division II). The species of 

 Division I all have the prosternal plate quite bare except in Neolimno- 

 phora. 



It is now necessary to examine the various proposed generic and 

 pubgeneric names and see how ihey fall within these two main divisions. 

 Taking them in chronological order, we have : — 



LiMNOPUOEA Dsv. 1880. — Desvoidy included fourteen supposed 

 European species in this genus, not one of which has been recognized 

 by subsequent authors ! He failed to specify any pai'ticular species as 

 tlie type of his genus, but Coquillett (1910) suggested that L. palustris 

 might be so considered ; and as Desvoidy appears to have used this species 

 as a sort of model with which he compared, or from which he built up, 

 many of his other species, there is much to be said in favour of 

 Coquillett's selection. Bvit what Avas L. 2)(ihisiris of Desvoidy ? One 

 naturally accepts the fact that it must have been a species answering to 

 Desvoidy's generic definition, and in this definition he calls attention 

 to a somewhat prominent " epistoma " — a term by which he indicated 

 tiie front mouth edge; to the fact that the dark abdominal markings in 

 the female are as large as in the male ; and to the colour of the species 

 boing " brunneo-cinerei." The first two characters apply only to species 

 of the division with base of the cubital vein and prosternal plate setose, 

 and the colour supj)orts the same conclusion. One is therefore almost 

 certainly correct in considering L. j^c^ustris to be a species allied to 

 L. maculosa, and in retaining the name Limnophora for a part of 

 Division II. It is interesting to note that though Schnabl did not 

 recognize the essential characters of the group, his limitation of the use 

 of the name Limnojjhora is almost the same as that now proposed. 



GrTMNODiA Dsv. (1863), with type ;)r«/(°ns/s Dsv.=^jo/y- 

 sf/(/ma Mg., belongs to Division I. The upcurved cubital vein led 

 Desvoidy to place the genus next to Graphomyia, with which it really 

 Ikis no relationship. The synonym}'^ of the type-species is gen< rally 

 accepted as correct, though Desvoidy's characterization of the genus was 

 " Caracteres des Graphomyes ; Chete nu " and imder Grapliomyia lie 

 called attention to the hairy eyes, leading one to infer that his Gi^mnoiUa 

 also had hairy eyes, whereas the e^'es oi poly stir/ ma are bare. 



Melanochelia Kond. (1866), with type surda Kond. {nee Zett.) 

 =^exsurda Pand., belongs to Division II. 



Brontaea Kow. (1878), with tA'pe polysiiyma Mg., belongs to 

 Division I. and is a sj-nonym of Gymnodia. 



