14 RK. We CROSSKEW 
never both in the same taxon). In some genus-group segregates the postalar wall 
(i.e. the vertical lateral declivity of the postalar callus, below the rounded margin 
bearing the setae) has a thick tuft of long dense hair, but in most groups the postalar 
wall is bare (or at most has just one or two hairs on its extreme upper part imme- 
diately below the ridge of the callus) ; in other segregates the suprasquamal ridge is 
most often haired, either with long dense bushy crinkled hair which is so thick that 
the centre part of the suprasquamal ridge cannot be seen or with rather short sparse 
hair under which the whole of the ridge is clearly visible (the nature of the hairing 
therefore provides a useful character as well as its presence). A few segregates 
(including the genus Rutilodexia, the subgenus Ameniamima and some species of 
Rutilia s.str.) have both the suprasquamal ridge and the postalar wall bare, but most 
often one or the other of these structures is hairy. 
Presence or absence of hair on the prosternal membrane (Text-fig. 18) has some 
taxonomic value. In Chrysorutilia and the aberrant species Rutilia micropalpis 
there is hair on the anterior margin of the prosternum as well as on the membrane, 
but hair actually on the prosternum itself does not occur in any other Rutiliini. 
The hairing of the abdomen and male hypopygium shows no characteristics of 
supraspecific taxonomic value, but there are sometimes minor differences at the 
specific level in the length, strength and bushiness of the hairing, especially on the 
epandrium and surstyli. Hairing on the surstyli provides some particularly good 
specific characters in Prodiaphania. 
Hairing on the arista ranges from extremely short micropubescence to moderately 
long plumosity, and is generally similar in the species of any particular subgenus; 
aristal hairing therefore provides a character of some minor value at the supra- 
specific level. 
Heap (Text-figs 1-3) 
The most important taxonomic character at supraspecific level is the shape of the 
facial carina, which Paramonov (1968 : 355) used in the first couplet of his key to 
Rutiliine genera. Particular genera and subgenera usually show a moderately 
constant facies in carina shape, but there is normally also some intraspecific and 
interspecific variation and the character is not so easy to use in practice as Para- 
monov’s key implies. In some segregates the carina forms a prominent convex 
knob between the antennal bases and becomes slender and sharper towards the 
epistome, while in others it is very broad along its length and has subparallel sides; 
the latter form of carina often shows, superimposed upon its basic shape, either a 
trace of a median sulcus or a median ridge. The efistome is always slightly promi- 
nent in profile, and often subnasute, and is rather constant in shape in any segregate; 
in forms with a bulbous facial carina the epistome is normally very prominent and 
the face in profile is deeply concave between carina and epistome, but in forms with a 
broad flattened carina the epistome is usually less prominent and not conspicuously 
differentiated in profile from the epistome by a deep saddle. Most Rutiliines have a 
well developed haired dilation on the gena (the so-called genal dilation), but in a few 
forms (especially Rutilodexia) there is very little genal dilation and the dilated part 

