92 MODERN CLASSIFICATION OF INSECTS. 
metathorax or metascutellum (fg. 72. 2. and 3. s.), is divided from 
its anterior part or metapraescutum (ibid. z.), by a deep impression, 
extending down the sides (incorrectly figured by Burmeister, Manual, 
pl. 12. No. 1. f. 2.), so that it appears to be a distinct segment (and 
consequently portion of the abdomen), it is also furnished on each side 
with a spiracle (fig. 72. 3. 0.).* 
* That this hinder division (s.) is reallya portion of the mesothorax may at once be 
seen by carefully comparing figures 72. 1, 2, 3. with the thorax of Cephus, Urocerus, or 
especially Oryssus ( fig. 73. 6,7-); in all which, as indeed in other shorter-bodied 
species, the dorsal and ventral ares of the real abdominal segments will be found to 
correspond together. In Cimbex ( jig. 72. 2, 3.), the episternum (ibid. vy. ), and epi- 
meron (ibid. y.) of the metathorax, are well defined, the latter giving support to 
the base of the posterior coxz (ibid. t.), whilst the metascutellum is large, with the 
edges acute, and the sides deflexed (ibid. s s.), and extending to the base of the 
posterior coxz, having the metathoracie spiracle (0. ) on its deflexed sides, The analogy 
of these pieces with those of Vespa, figured by MacLeay (Zool. Journ. vol. v. pl. 5. 
fig. 1.) is unquestionable. In Oryssus, the metathoracie epimeron and episternum 
are confluent (in the piece marked y. in fig. 73. 7.), and the metaseutellum (ibid. s.) 
has no acutely deflexed side, having a longitudinal direction. It might, perhaps, 
be at first supposed that the two ares r. in this figure represented ss. in fig. 72. 3.3 
but the position of the Cenchri ( x ) in the piece marked z. in fig. 73. 7. ; and the 
circumstance that these two ares haye a distinet motion like the other abdominal seg- 
ments, clearly proves that s. is‘ metathoracie and r. abdominal. In some species 
(Cimbex, jig. 72. 1, 2, 3.), the posterior margin of the metathorax is deeply emar- 
ginate, so that it is connected with the first abdominal dorsal are by a slender 
white membrane (jig. 72. 2, 3.+), which I regard as analogous to the funiculus K. 
of the petiolated species. In many other species, as in Allantus, Tenthredo (and 
also in Urocerus), this hinder part of the metathorax is slit down the centre, whilst 
in Cephus, &c., the slit is dilated into a triangular excision, covered with yellow 
membrane. Great confusion and consequent inaccuracy in the description of the 
colouring of the respective abdominal segments has originated in the want of a care- 
ful discrimination of the structure of this piece ; for instance, in Allantus lividus the 
basal segment of the abdomen is said to have a white spot on each side (Steph. Til. 
Mand. vii. p. 66.), whereas it is the posterior portion of the metathorax which is 
thus marked. In Zarza fasciata, 2, the colours are still more strongly marked. By 
all authors the abdomen is described as black, with the basal segment white, whereas 
the white segment is part of the metathorax. In Oryssus coronatus the abdomen is 
said by all authors to be rufous, except the two black basal joints; whereas it is the 
hinder portion of the metathorax (which is not slit, fig. 73. 6...) and the basal seg- 
ment of the abdomen alone (r.) which are coloured black. In Urocerus, likewise, the ab- 
domen is described as 9-jointed, whereas it has but eight segments; the hinder part 
of the metathorax having been regarded as the basal segment of the abdomen. The 
notes of Latreille upon this subject are completely at variance, supporting two 
distinct theories. He distinctly states (Régne An. tom. v. p. 268.), that the seg- 
ment which bears the inferior wings is separated from the following, “ou du premier 
de abdomen, par une incision ou articulation transverse.” Now the part which pre- 
