412 C. GORDON UEWITT. 



{B. 2+3). This cuts off the first radial cell (1 li.). The last 

 nervure, which joins the costal almost at the apex of the 

 wing, represents the fused fourth and fifth radial nervures 

 {E. 4 + 5), and so cuts off the third radial cell (3 7?.). The 

 fourth main longitudinal nervure is the median^ which, in the 

 typical insect, divides into three, but in the fly the nervui-es 

 have undergone coalescence, as will be shown. The first and 

 second median nervures have coalesced {31. 1+2), and do 

 not run direct to the margin of the wing, but bend, forwards 

 and almost meet i?. 4 + 5 on the costa. About half way across 

 the wing a transverse nervure, the radio-medial {rm.) unites 

 jB. 4 + 5 and if. 1 + 2, and cuts off the fifth radial cell (5 B.) 

 from the radial (Z?.). The next longitudinal nervui'e repre- 

 sents the coalesced third medial and cubital nervures {M. 

 3 + Ow. 1). It runs to the posterior margin of the wing 

 about half way along the length of the latter. The nervures 

 31. 1+2 and 31. 3 + Cu. 1 are united by two transverse 

 nervures. The proximal nervure — the niedio-cubital {m.cu.) 

 cuts off the small triangular medial cell {31.) ; the distal trans- 

 verse nervure {m.) cuts off the first second medial cell (2 31.^) 

 from the second second medial cell (2 il/.^). The last longi- 

 tudinal nervure — the anal {A-^.) — is undivided, and does not 

 reach the margin of the wing, thus incompletely separating 

 the first cubital (1 Cii.) and anal (^4.) cells. A small trans- 

 verse nervure, the cubito-anal {cii.a.), slightly more proximal 

 than the medio-cubital, cuts off the small triangular cubital 

 cell {Cu.) from the first cubital cell (1 Cu). Running parallel 

 with, and posterior to, the anal longitudinal nervure, tliei'C is 

 apparently another nervure. This, however, is not a true 

 nervure, but is merely a chitinised furrow giving additional 

 strength to the posterior angle of the wing. The posterior 

 edge of the base of the wing is divided into a number of 

 lobes. These are the anal lobe, and, as Sharp (1895) pro- 

 posed, the alula, antisquama, and squama. The squama is 

 thicker than the rest, and is attached posteriorly to the wing 

 root between the mesoscutum and the lateral plates of the 



