50 REPORT 4, UNITED STATES ENTOMOLOGICAL COMMISSION. 
air- chambers in the interior, accompanied by manifold ingrowths of the 
outer crust to serve as partitions between the adjacent air-chambers. 
As the ingrowths are connected with divisions of the exoskeleton, this 
last becomes very complicated, especially in the posterior lateral region 
below the wing and above the leg. The scutum, s 3 , is well marked; 
immediately below it arises the hind wing, w 2 . The piece which Bur- 
gess determined as the scutellum in Danais we have not found in Ale- 
tia ; but what in the butterfly appears merely as the tip of the scutum is 
distinctly differentiated in the moth and is very probably the true scutel- 
lum; in this case the part so named in Danais would have to be con- 
sidered as the post-scutellum. On the front edge of the segment, be- 
tween the front edge of the wing and the coxal joint, is a single piece, 
eps 3 , which seems to correspond to the two pieces, eps 2 and st 2 9 of the 
mesothorax, fused into one. In the posterior part, epm 3 , there are a 
variety of structures, of which the most important are two, marked A 
and jB. The former is a little quadrangular flap, which hangs down 
from just below the posterior edge of the hind wings. The latter, B, 
is a deep-lying, oval, pellucid membrane, which we think is probably 
homologous with the tympanal membrane of grasshoppers. The rela- 
tion of these parts is better shown in the enlarged figure, Plate XI, 
Fig. 3, in which the oval membrane, I>, and the flap, J., are both very dis- 
tinctly drawn. We have not succeeded in observing any spiracle on 
this segment. The boundary between the metathorax and the abdomen 
is not clearly marked externally. The coxal joint, c# 3 , is similar to that 
of the second leg. 
A. H. Swinton has published a paper on the organ of hearing in lep- 
idoptera,* in which he refers to the oval disk, which we have interpreted 
as a tympanum. According to Swinton, a nerve passes from the third 
thoracic ganglion obliquely across and round the elevator muscle of 
the hind wing to the supposed tympanum, where it is connected with a 
structure (Swinton's " membranous vesicle") which is apparently iden- 
tical with the structure in like position in the grasshoppers, which latter 
was likewise originally described as a vesicle, but is now known to 
be really a cluster of rod-bearing, terminal organs, such as are now 
known to be the essential constituents of tympanal organs. For a gen- 
eral account of these apparatus see the r6suin6 by C. S. Minot.t It is 
probable that the part we have described in moths is a real tj mpanum, 
and entirely homologous with that of Acridians, but the matter must 
remain uncertain until the terminal rods have been actually found. 
The abdomen consists of nine segments, numbered 1-9, the last two 
not showing in the figure, being retracted into the seventh segment. 
The first is smaller than the second and succeeding segments, and, 
therefore, appears as a sort of thick stalk uniting the abdomen with 
* Swinton, A. H. On an Organ of Hearing in Insects, with special reference to the Lcpidoptera. En- 
tomologist's Monthly Mag., XIV (1877), 121-126. 
t Minot, C. S. Comparative Morphology of the Ear. Fourth Article. American J ourn. Otology, IV 
(1882), 80-168. 
