50 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



fiimily is a composite or synthetic one, embracing- forms both leading- up to and including- the 

 Sphingidie, all bound together by genetic ties. As there are already too many moditications of 

 the names Sphingidw and Bombycidffi, we venture to hope that the name we here propose 

 ma}' be accepted by entomologists. For all the syssphingine families, below Sphingidse, I pro- 

 pose the name Prvtosplihigina. 



XIII. ORIGIN OF THE SYSSPHINGINA BY BOTH CONTINUOUS AND DIS- 

 CONTINUOUS EVOLUTION. 



The results of our studies have taught us that two modes of evolution have been at work in 

 the origin of the family of Sphingidiv. First, there was, due to a change of habits, a gradual, 

 continuous process of progressive modification of the small-headed, short-tongued, thick-bodied, 

 sluggish, or nearlj- flightless Saturniides by way of the Ceratocampinffi into a Cressonia-like 

 form. This process of change and adaptation to new conditions of life went on for perhaps 

 many centuries or thousands of generations. At length there was a sudden acceleration or 

 revival of growth and development in those partly atrophied organs like the maxilhv, etc., which 

 became restored to the functions enjoyed by the more active ancestors of the saturnian 

 subphylum, and at a critical period, after one consisting of long preparatory but slight changes, 

 a per saltmn movement or leap occurred, and as the result of this rapid assumption of a new 

 character, which we call an aberration, sport, or mutation, the tul)ercle r of the larvse became 

 shifted fronl its position in the cei-atocampid larvae to what it is in Cressonia, Ceratomia. and the 

 other Sphingida thus far examined, and there also appeared an additional radial vein. 



This has been the case with the origin of the genera not only of this group, but this process 

 of frequent rapid evolution takes place in the organic world in general. An example is the 

 differences between Adelocephala and Anisota, the larva? of the latter genus differing so 

 remarkabi}' from those of the stem-form in the return to the primitive separate tubercles ii of 

 the eighth abdominal segment and the reduction in the armature, that only a single pair of 

 thoracic horns are left. 



Doctor Dyar " has called attention to the discontinuous evolution of a wart bearing hairs 

 from a simple setiferous tubercle, stating that " we do not find a series of intergrading forms 

 ))etween the single-haired tubercle and the many-haired wart, though both may occur in different 

 genera of the same family." The wart-like tubercles which characterize the Saturniidre are 

 apparently suddenly produced characters; also the peculiar branched tul)ercle spines of the larval 

 Hemileucida>, and certainly the lateral eversible glands which are peculiar to and diagnostic of 

 that family. In fact the fusion of the two tubercles i on the eighth abdominal segment of the 

 Syssphingina is a case of more or less sudden or rapid evolution. Thus the discovery that Bom- 

 bycidic {Bomhyx viori), Brahmaddi^?, and Endromis versicolora (Endromida?) all have in the first 

 larval stage warts bearing several hairs proves that thej^ belong to a different phj'lum from the 

 S.yssphingina and should be associated with what we would call Si/inhohJiycina, including- the 

 Eupterotida% LasiocampidiB, and Liparida\ these families, perhaps, having arisen from the uoto- 

 doiitian groups Apatelodina? and Ichthyurina' (Melalophinw). 



We hence infer that those absolute characters which distinguish or are diagnostic of lepi- 

 dopterous families, however slight or trivial in themselves, are sudden acquisitions, due perhaps 

 to compai-atively sudden changes in the conditions of life, involving new needs, the formation of 

 new habits, different food plants, etc., or some unknown stimulus. If this be the case, then the 

 different family groups, as well as generic groups in the Lepidoptera, have arisen as sudden 

 departures or changes or divergent steps in the course of what otherwise would be a slow, evenly 

 graduated process of progressive development. If this is the case with Lcpidoj^tera so it is in 

 other orders of insects and other arthropodous phyla, and, indeed, throughout the organic world. 

 For example, the birds with more or less suddenness diverged from the reptilian line of descent; 

 mammals with two condyles originated l)y a process of rapid evolution from reptiles with Imt 

 one condyle, and so on. 



« Annals N. Y. Acad. Sc, viii, May, 1894, p. 196. 



