60 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



primitive species of Adelocephala hi<rh and of equal size. In Auisota but a single pair of dorsal 

 niesothoracic horns, and the caudal horn is represented by two small separate tubercles. 



In color and armature protected bj^ their resemblance to the color of the foliage and .spines of 

 the spiny plants on which the more primitive species feed. 



In stage I the body bears no glandular hairs, and the thoracic horns and caudal horn in all 

 except Anisota are enormously long and end in a more or less bulbous knob, which gives rise to 

 two rod-like seta?. 



Piqxi. — The head is unusually full and rounded, overhanging the mouth-region and, the pupa 

 being subterranean, is not provided with any large protuberances; the mouth parts present no 

 vestigial characters; the maxillae vary in length, the two together in Adelocephala and Anisota, 

 whose imagines have nearly aborted tongues being scarcely longer than bi'oad; abdominal sutures 

 deeply impressed. The armature consists of small sharp spines, either on the head between the 

 antenna? or on the prothorax, and elsewhere on the thorax. Metathoracic segment with a basal 

 transverse irregular ridge on each side; the cremaster is very long, flattened, and deeply forked, 

 or in Citheronia it may be nearly atrophied and functionless; the a))dominal segments usuallj' 

 with two rows of fine shai'p spines, but in Citheronia the segments are smooth. The surface of 

 the body difl'ers in degree of rugositj', either being deeply punctured or (Cithei'onia) nearly 

 smooth. 



The pupa?, then, of this group are recognized by the full rounded head-region and the 

 unusually long, forked, flat cremaster. 



History of the group. — As observed l)v Walsh, the group Ceratocampina^, as here understood, 

 was flrst established as a distinct and exclusive American family by Doctor Harris, ^ but has 

 been somewhat unnaturally united with Saturniadaj b^' succeeding authors," referring to Harris, 

 himself (Inj. Ins., p. 89S), and to Morris. He adds that "Doctor Clemens has beautifully shown 

 that it difl'ers from Saturniadw, not onl\- in the characters laid down by Doctor Harris, but also in 

 having the subcosto-inferior nervule of the front wing simple and not furcate. (Proc. Ent. 

 Soc. Phil., i, p. 177.) In other words, in Ceratocampada^ the subcostal vein sends oft' toward its 

 tip three branches or sectors which are all simple; in Saturniida', the middle one of these three 

 sectors, instead of rising directly from the subco.stal vein, rises from the basal sector or branch, 

 thus making that basal sector appear furcate.'''' (Proc. Boston Soc. Nat. Hist., ix, Feb.. 1S04, 

 p. 2(t(t. footnote.) 



B3' Grote (List Lep. , ISOS) the name Ceratocampinse was proposed for the Ceratocampidae 

 of Harris and the present autiior, Grote then regarding it as a subfamily of Pombycidw. 



The name Ceratocampina? was also given to the group by J. B. Smith in 1SS6, it being hy 

 him regarded as a subfamily of Saturniida;. 



It received the family name Citheroniidic from Dyar in ISlMi, who proposed that name on 

 account of the generic name Ceratocampa being a synonym; in 1902 the same author restored 

 the more fitting name Ceratocampidic. In 1901 (Psyche, ix, p. 280), however, we showed that 

 it is a subdivision of the now more comprehensive family group for which we would use the 

 name Ceratocampidae. 



As it now stands the family Ceratocampida; of the older authors, Harris and others, has been 

 found to be a subdivision of a much more extensive group of family, and possibly superfamily 

 rank. The family Ceratocampid;v as I have enlarged it, and as described in the foregoing pages, 

 differs in ^gg., larval, pupal, and adult characters from either the Hemileucidte or the Saturniida. 

 None of the members are spinners, but when about to transform the larva enters the earth, the 

 pupa being subterranean, its cremaster being a large stout spine. 



To this extensive family group belong the great moths of central and southern Africa 

 (Afroga?a), Mhicli have iiitherto been regarded as true Saturniida'. For example. Nudaurelia, of 

 which there are about twenty species, and which, until separated by Rothschild, were confounded 

 with the Asiatic genus Anthertea, is in it^ larval, pupal, and imaginal characters closely allied to 

 our American Ceratocampina\ This is also the case with (iini<(n!sa isi.s, etc. 



As the result of prolonged study of the venation, and what little we know of the larval and 

 pupal characters, it is necessary to remove many, indeed most, of the African genera heretofore 



