283 
stances erroneous, so that I assumed species with eight-jointed 
antennae to be distinct from those which Sir William asserted 
to have nine antennal joints. After many attempts to identify 
Macleay's species by the study of their author's descriptions I 
arrived at the conclusion that it was impossible to do so, and 
that an examination of the types (which are in the Sydney 
Museums) was essential. Accordingly I have recently visited 
Sydney for the purpose of making that examination, and am 
now in a jDosition to deal with the matter authoritatively, 
and the result of my investigations will be found in the fol- 
lowing pages. 
The number of names that I can ascertain to have been 
given to species presumably of JAyaretrii^, is 130, of whicn 
29 may be confidently regarded as synonyms, and 9 are so 
described that they cannot be identified without the examina- 
tion of types to which I have no means of access, the number 
of recognisable species being, therefore, 92. To these I have 
now to add 20 new species, bringing the total up to 112. 
As remarked above, the species of Liparetrus are in 
general distinguished by good structural characters, which on 
first thoughts would suggest the probability of its being easy 
to break the genus up into satisfactory subgenera and sec- 
tions ; but a prolonged and careful study has forced me to the 
conclusion that the structural differences are so curiously 
intermingled that there is not one of them by means of which 
anything approaching a natural group can be formed. The 
most striking of the structural characters that I refer to are — 
(a) nature of sexual differences: fh) form of clypeus : (c) 
structure of hind tarsi ; ( dj vestiture of dorsal surface ; { e) 
structure of front tibiae ; ff) structure of antennae 
The species, however, which are associated by agreement in 
any one of these respects differ \videly as regards the other 
respects, and the species which are placed together by reli- 
ance upon any of them are not naturally associated, and have 
their closest allies in other groups. I have, therefore, not 
thought it well to form any subgenera, but have made ths 
best use I can of the structural characters for grouping, 
without claiming to have succeeded in accomplishing a break- 
ing up into natural aggregates except in so far as I shall 
indicate in the course of this paper that one or two of the 
subordinate aggregates seem to be a natural association of 
species. 
It must be noted here, however, that J.ipnretj'ns, as 
treated by Blanchard and Macleay, includes a number of 
species of a genus separated by Burmeister from Liparetrus 
under the name Autonwhis (which I hope to discuss in a 
future memoir under the heading of that name), and it is to 
