560 Notes in 1887 upon lepidopterous larve, éc. 
The five spines are covered with large thorn-like pro- 
cesses, each of which emits a single bristle, while 
especially large bristles project from the extremities of 
each prong of the terminal fork. The structure is 
therefore essentially similar to that of the caudal horn 
in all young Sphingide hitherto described. As far as 
the thorny appearance is concerned, the resemblance is 
to the caudal horn of the second and third stages of 
Sphinx rather than the first stage, while the bifurcation 
of Aglia is chiefly recalled by the condition of the first 
stage of Sphingide, in which it is most marked. These 
differences in the condition of the horn in successive 
stages have, however, only been carefully studied in 
S. convolvuli among Sphinx larve, although there is no 
reason to doubt their occurrence elsewhere (compare 
Plate XYV., figs. 3, 9, and 16, with Plate XVIL., 
fig. 4, x 50). The five spines of Aglia are bright red 
(like the caudal horn of Smerinthus ocellatus in the three 
first stages), except for one section of their length, in 
which the pigment is absent, and the structure is white 
and comparatively transparent. This section is situated 
nearer to the fork than the base of the spines (see 
Plate XVII., figs. 3 and 4). I was greatly interested to 
find even this character represented in the caudal horn 
of the 2nd, 3rd, and 4th stages of Sphinx convolvuli, 
which is deeply pigmented (in this case with black) at 
the tip and base, but of which a short section, nearer to 
the former, is white and more or less transparent (see 
Plate XV., figs. 9 and 16, &c.). 
When the larve are irritated they move their spines, 
especially the 3rd thoracic pair. The usual position of 
the spines during rest is shown in Plate XVIL., fig. 3. 
It will be remembered that the caudal horn of the 
larval Sphinx ligustri in the 1st stage is also movable 
(Trans. Ent. Soc. Lond., 1885, p. 288). 
In addition to the tubercles already described, and the 
spines which represent certain of these structures, there 
are also other similar tubercles on the thoracic seg- 
ments, and on the abdominal segments posterior to the 
7th. The appearance and arrangement of these is to 
some extent shown in fig. 3, but peculiar interest 
attaches to the two pairs of tubercles between the caudal 
horn and the spine on the anal flap (anal spine). 
Entirely independent investigations (upon the |lepi- 
dopterous pupa) have long since convinced me that 
