136 



cidce. Attacus luna and Polyphemus^ notwithstanding the 

 difference in their hind wings, cannot be separated generically, 

 for their larvas [Ph iv, figs. 14, 17] and cocoons are ahnost ex- 

 actly ahke. These two will form one section of the genus, 

 while Cecropia and Promethea constitute another. Bomhyx lo 

 is much nearer to Saturnia than is Proserpina^ inasmuch as 

 the joints of the antennae are each doubly bipectinated. In 

 Afjlia tau each joint seems to have only one branch on each 

 side ; but there are, in fact, two branches which are so close 

 together as to appear like one. In our Proserpina I cannot 

 find that the apparently single branches are in reality bifid ; 

 nevertheless Proserpina comes nearer to Aglia than to Satur- 

 nia. Perhaps lo and Proserp)ina should be types of new 

 genera. Endromis versicolor^ in the perfect state, is evidently 

 very near to our Proserpina ; but is, with great propriety, made 

 the type of a distinct genus, on account of the form and habit 

 of the larva. It approximates very remotely to Bomhyx mori, 

 to use an Irishism. The Ceratocampiadoi cannot come among the 

 Notodo7itiadai (although like some of the latter, e. (j.,Pygcera, 

 they are not true spinners), because the hind wings of the 

 former are reversed in repose, in which they exactly agree with 

 B. lo ; while the hind wings in the Notodontiadce are never re- 

 versed. The cocoon of B. lo is very slight, consisting only of 

 a little gummy matter, with which a few leaves are fastened 

 together and lined. B. Proserpina is said l)y Abbot to trans- 

 form in the ground without making a cocoon ; both of them are 

 gregarious, at least while young. The Ceratocampiadoi are gre- 

 garious also, and transform in the earth without making cocoons. 

 In this respect, too, they agree with Pygcera and the larva of 

 Bryocampa senatoria. Beginning the Notodontiadm with Py- 

 gcsra, I leave it with Limacodes, which certainly ought to come 

 in this family, rather than among the Arctirn. The cocoons 

 are egg-shaped, and very "hard, as those of Cerura. Without 

 doubt, that very curious and beautiful moth, the B. torrefacta 

 of Smith- Abbot, should come among the Liparidce. It is the 



