60 TRANSACTIONS OF THE [Nov. 12, 
horns of Ceratomia amyntor might be homologous with those 
of the Citheroniidz, I can show to be unfounded, for in the first 
larval stage of this Sphinx these processes arise anterior to the 
sete and entirely independently of them, whereas in the Cither- 
oniide the horns are developed out of the corresponding tuber- 
cles. (See Fig. 4. Tubercles iii and v are absent (as usual in 
stage I.) and in Ceratomia vi is represented by two sete. At 
a is shown the rudiment of the future thoracic horn of Cera- 
tomia.) 
Superfamily RuopALoceRA. 
Tubercles and sete primitive, often rudimentary and disap- 
pearing at first moult. Tubercles iv and v usually in line, 
rarely iv moved up a little or considerably out of line (Danaine, 
Nymphaline). Armature processes of mature larva, when pres- 
ent, derived independently of the tubercles, but often occupying 
approximately the same position. Includes the families Papili- 
onide, Pieride, Hesperidee, Lyceenidz and Nymphalide. 
The setz are either single or multiple in the first stage. The 
multiplication is very slight in some cases, so that it is difficult 
to say at what point the primitive first stage has disappeared. 
It is definitely lacking in most of the Papilionide, though this 
family is otherwise among the most generalized. ‘The processes 
of the Nymphalidz do not appear till after the first moult or 
later. and the unpaired dorsal series of this family is independ- 
ently derived and does not correspond to the unpaired tubercles 
of the Saturnina.* 
*The figure of Melitea phaeton. stage I, by Dr. A. Gruber (Jena Zeits. fiir Natur- 
wissenschaft, Vol. XVIII: see reprint in Papilio, Vol. LV, pl. iii, Fig. 25) appears to con- 
tradict my generalization. It represents a single dorsal seta with no sign of consolida- 
tion. I have not seen Melitza, but Mr. S. H. Scudder (‘‘ Butterflies of the Eastern U.S. 
and Canada, Vol. I, pp. 619 and 688) gives careful descriptions of it. He says, ‘‘ Papillee 
all arranged in paired rows,’ and further describes the arrangement in detail. 
It is quite normal, except that iv. seems to be moved up a little. 
Mr. Scudder devotes four plates to figures of newly hatched larvee. In most of these 
figures the sete are very plainly shown, and iy. and y. are approximately in line. Plate 
70, Fig. 3, well illustrates the peculiar arrangement of Danais archippus, showing the 
additional subdorsal seta above i. Vanessa antiopa (pl. 70, Fig. 12) shows: tubercle iy. 
moved up behind the spiracle, very much as in the Noctuina, and Grapta faunus (Fig. 
8) issimilar, apparently with the addition of tubercle vi., bearing two sete. It thus ap- 
pears that in certain genera of the Nymphalide (the most specialized family) the primi- 
tive arrangement of the sete is displaced, perhaps owing to a tendency to the disappear- 
ance of the primitive first stage. In Grapta progne, which I have examined, iy, is only 
moved up a very little out of line with y., but v. bears two setze. Mr. Scudder’s figures of 
the Lycznide (pl. 71) show a number of supplementary sete, the most generalized con- 
dition being shown in Fig. 2(Thecla liparops), where the arrangement seems normal. 
The figure of Pieris rape (pl. 72, Fig. 8) is poor. The arrangement of the sete is really 
normal, iv. and y. being in line. In Anthrocharis genutia (pl. 73, Fig. 9) tuberele y. 
seems to be wanting. 
_ In the Papilionidee, more than in any of the other families, the warts are derived 
. from the primitive tubercles. The correspondence with the moths is close, but the 
single wart below the spiracles, though corresponding to iy. and y., does not seem to be 
