370 MODERN CLASSIFICATION OF INSECTS. 
moths, and some of these seem so nearly related to Coronis, whilst 
Urania is in several respects so close to Agarista (in its larva, palpi, 
and antenne), that Iam induced to unite them. into one family, a 
step which seems to be supported by the neuration of the wings. 
Even between Castnia and Urania, Mr. MacLeay has no doubt that 
there exists ‘some close kind of relation.” (Trans. Zool. Soc. vol. i. 
p- 188.) How far their relation to the Anthroceridze by means of 
their transformations is to be considered as closer than to the Hes- 
periide, remains for future consideration. I cannot, however, but add 
that Urania appears in some respects to approach Erebus, especially 
in the palpi*, and Mr. MacLeay notices the resemblance between the 
eggs of Urania and Catocala, to which last genus also Castnia must 
be allied. The antennz are long, but never prismatic as in the 
Sphingide, nor terminated by a pencil of hairs, but are variable in 
form; in Urania they are slender, filiform, and gradually attenuated 
to the tip; in Coronis and Castnia, however, they are clavate at the 
tip; the maxilla are elongated ; the palpi are distinctly 3-jointed; in 
Urania and Agarista they are elongated, the second joint very com- 
pressed, and the third slender and naked ; the fore tibiz are spurred 
(fig. 103. 2.). 
The accounts published of the transformations of Urania (Cy- 
dimon Dalm., Leilus Swains.) have varied considerably ; Madame 
Merian having described a larva and imago from Surinam supposed 
to be those of Urania Leilus, the former armed with very long and 
strong spines as hard as iron wire. M. Sganzin, on the other hand, 
asserts that the larva of Urania Rhipheus (Thaliura Rh. Duncan, 
For. Butt. pl. 28.) is a semi-looper, and its chrysalis naked, sus- 
pended by the tail and girt round the centre. (Anomalie du g. 
Uranie par Boisduval, Ann. Soc. Ent. Franc. 1834, p. 248., and 
Nouv. Ann. du Mus. p. 260.) But the elaborate account pub- 
lished by Mr. MacLeay upon the habits and transformations of 
Urania Fernandine, proves that its larva (jig. 103. 1.) closely 
resembles that of Agarista, being cylindrical with long delicate 
setae, and that the pupa as in that genus is enclosed in a cocoon, 
but of so Jax a nature as to allow the inmate to be seen. Mr. Mac- 
Leay does not mention whether the chrysalis is attached by its tail 
as in the butterflies; but since his return from Cuba he has been so 
* T find that Fabricius had placed some of the Uranie with Erebus, thus an- 
ticipating my supposition of this relation. 
