16 R. W. CROSSKEY 
be present in reduced form), and it appears that this might provide a useful specific 
character. There are only minute differences in the relative proportions of the 
costal sectors, but the degree of dilation of the costal base provides a character of 
minor taxonomic use; some species have the costal base exceptionally strongly 
flattened and widened (explanate), especially in the male, and the explanate costal 
bases (i.e. the antecosta of Paramonov) then give the wings the appearance of having 
basal ‘shoulders’ which are very easily visible to the naked eye (especially in flies 
with the wings set back in the resting position). Partial or complete infuscation of 
the wings occurs in some species, and can provide a helpful specific character. 
Basicostal colour is also sometimes helpful. 
ABDOMEN AND GENITALIA 
Abdominal form. The abdomen is always very robust (Text-figs 28 & 29) and 
often conspicuously broad and slightly flattened, and the shape is moderately 
uniform in each genus-group segregate. The shape of T5—if convexly rounded and 
tapering posteriorly or whether short, very broad and deeply hollowed medially— 
provides an important character at supraspecific level. The degree to which the 
sternites are exposed between the approximated ventral ends of the tergites does not 
seem to provide any helpful characters except at the specific level in Prodiaphania. 
The bilobed fifth sternite of the male shows little variation through the tribe as a 
whole (Text-figs 30-35), the two lobes normally having a simple rounded shape 
(Text-figs 30 & 31), but an unusual shape of fifth sternite occurs in Grapholostylum 
(Text-fig. 33) and the lobes are straighter than usual apically in Microrutilia (Text- 
fig. 32). In the subgenus Paramphibolia each lobe has a small blunt tooth or prong 
on the inner edge towards the apex (Text-fig. 35), and this form of male fifth sternite 
occurs nowhere else in the Rutiliini. 
Male genitalia (Text-figs 36 & 39). The general structure of the male hypopygium 
is exceedingly constant throughout the Rutiliini and there are really no characters of 
importance at the supraspecific level; even at the specific level the taxonomic charac- 
ters are almost confined to the surstyli, with a few useful characters in the cerci and 
epandrium. The hypandrium (sternite 9), pregonites, postgonites and epiphallus — 
(spinus) are virtually identical in every species, and the aedeagus itself has an 
astonishingly constant form. The only difference observed in the aedeagus between 
different species lies in the distiphallus, in which the relative lengths of the proximal 
sclerotized part and the membranous distal part vary slightly. In the majority 
of forms the membranous part of the distiphallus is about as long as the sclerotized 
part or a little shorter (Text-fig. 37), but in a very few species (mainly the subgenus 
Grapholostylum) the distal membranous part is unusually elongate and whip-like 
(nearing twice as long or so as the sclerotized length, as in Text-fig. 38). The 
epandrium (To) is very large and shows some very minor differences in shape between 
species, but the differences are rather too intangible to have practical value in taxo- 
nomy. The cerci are elongate, not fused, and sometimes show differences in shape 
either in profile or posterior view which are useful for species recognition; they have © 
no features of use for supraspecific characters. The most important structures by — 

