60 R. W. GCROSSEKE Yi 
he cited no characters for it, but based it upon the nominal species R. formosa R.-D., 
for which a neotype is herein designated. As Townsend (1915) cited no characters 
of formosa it is herein presumed that no conflict exists between formosa R.-D. as 
here fixed by neotype (which equals formosa in Malloch’s and Enderlein’s sense) 
and Townsend’s oviginal meaning of this name. For further discussion of this see 
under neotype designation for R. formosa on p. 124. 
The limits of species in this group are often very uncertain. Some apparently 
distinct species which possess external pattern differences or differences of hair colour 
have male genitalia that differ but slightly or not at all. On the other hand, some 
very distinctive shapes occur in the male genitalia, especially in the shape of the 
surstyli, and where unusual surstylus shapes occur they appear to be constant and to 
define discrete species (for example, the bizarre shape of the surstylus in R. cryptica 
sp.n. is unmistakable and constant). In the past species have usually been founded 
on observed differences in external colour and pattern without regard to genital 
characters, and names have been bestowed on supposed ‘species’ in which the 
abdomen shows continuous transverse bands across the tergites and other names 
given to ‘species’ in which the tergite pattern is formed of completely or partially 
isolated spots. It has been found in the course of the present work that in many of 
the Chrysorutilia species there is little or no correlation between the male genital 
characters and the presence or absence or banded or spotted patterns. By removing 
the genitalia of a large number of male specimens of R. (C.) spblendida (as this species 
was previously understood) it has been found that three species have been confused 
under this name which are easily recognized by constantly different shapes of the 
male surstyli: in the commonest one (for which the name sflendida is fixed by 
neotype) the surstylus is simple in lateral and posterior view, having no trace 
of an anteromedian process (Text-fig. 72); in the second species (for which the name 
decora is fixed by neotype) the surstylus has a characteristic blunt swelling antero- 
medially which is just visible in posterior view (Text-fig. 73) ; and in the third species 
(which in the absence of an available name is here newly described as cryptica) the 
surstylus is produced anteromedially into an enormous forwardly directed tooth 
and is strongly excavate and acuminate between this tooth and its apex (Text-fig. 74) 
(the large tooth is very conspicuous in posterior view also). There is only a weak 
correlation in these three species between the genital form and the spot-pattern or 
band-pattern of the abdomen: in splendida the abdomen is usually banded but may 
have the bands broken into discrete spots, in decora the abdomen usually has discrete 
spots but may occasionally have continuous bands on one or both of the intermediate 
tergites, and in cryptica the pattern is similar to decora (most often spotted but banded 
specimens occurring). 
Another example of more than one species confused under a single name is that of 
R. (C.) imperialis. When males of this ‘species’ were examined for the present work 
it was found that their genitalia had two distinct forms: in one the surstylus has a 
blunt anteromedian projection (similar to that of decora) and the cerci are rather 
slender in profile with the tips bent slightly forwards (Text-fig. 75); in the other the 
surstylus is much narrower and has no anteromedian projection, and the cerci in 
profile are broad medially with a rather sudden contraction before the rather straight 

