

REVISIONARY CLASSIFICATION OF RUTILIINI 71 
tibia, both have extremely bushy and often crinkled hair on the suprasquamal 
ridge and there is normally only a single (posterior) sternopleural seta; other 
resemblances include the occurrence in most forms of hairing on the prosternal 
membrane, and general similarity in the shape of the facial carina (which tends to be 
rather flattened on the anterior face and to merge rather gradually into the lunula 
so that there is only a weak depression where carina and lunula coalesce). The 
broad sulcate last abdominal tergite with its transverse row of strong setae provides 
the most obvious external feature distinguishing Donovanius from Chrysorutilia, 
though the difference in the pteropleural hairing mentioned in the subgeneric key 
and shown in Text-figs 19 & 20 appears to apply constantly throughout the sub- 
genera and to provide a real distinction. The large heavy foliaceous surstyli of the J 
genitalia of Donovanius are very different from the narrow pointed surstyli found in 
Chrysorutilia (cf. Text-figs 66-71 & 72-84), and the g genitalia therefore readily 
distinguish these two subgenera. Another minor difference between them, but one 
which is apparently constant, is the lack of any hairing actually on the prosternum in 
Donovanius which contrasts with the presence of at least a few fine hairs on the 
anterior corners of the prosternum in Chrysorutilia. 
The species of Rutilia s.str. look superficially much like Donovanius on account 
_ of the grooved apex to the abdomen, but are at once separable by the presence of 
only three postalar setae and by the lack of a hind tibial fringe, and by the differently 
shaped ¢ surstyli (cf. Text-figs 54-57 & 66-71). 
Donovanius species appear to be entirely unrepresented in the Oriental Region, 
but the distribution of the subgenus is more extensive in the Pacific islands than that 
of other subgenera, which are either confined to Australia (Neorutilia, Ameniamima, 
Rutilia s.str.) or else do not occur so far is known anywhere further east than New 
Guinea and Australia (Grapholostylum) or the New Hebrides (Chrysorutilia). One 
species of Donovanius is known from Fiji (tvansfuga Bezzi) and two from Samoa 
(savatiensis Malloch and nigrihivta Malloch), whence these are, respectively, the only 
Rutiliines known to occur. Samoa in Polynesia represents the easternmost limit 
_ of distribution of the tribe Rutiliini as a whole. 
The British Museum (Natural History) collection contains a female specimen of 
the Samoan species Rutilia (Donovanius) nigrihirta Malloch which was reared from a 
larva of a Lucanid beetle identified as a species of Aegus Macleay, probably A. 
wpoluensis Arrow; this seems to be the first fully authenticated host record for a 
species of subgenus Donovanius. 
The species of Donovanius do not aggregate in any obvious way, and no species- 
_ groups are recognized within the subgenus. The ¢ genitalia with their very distinc- 
| tive form of surstyli are extraordinarily uniform in all the species (or supposed species) 
of the subgenus. 
Most of the synonymies shown in the list of included species that follows require 
no special comment, but amplification is needed here concerning R. (D.) sabrata 
(Walker) and R. (D.) bisetosa (Enderlein). One of the species of Donovanius is a 
large blackish brown form with dark violet reflections and unusually long antennae 
compared to those of other Rutilia species, and it was to this species that Malloch 
(1927 : 347; 1929 : 300) applied Guérin-Méneville’s name inornata, using (in the 
