MEMOIRS or THE NATIONAL ACADEMY OF SCIENCES. 27 



grooves passing down tlie shaft from the miti-hes between the teeth. They occur not only on the 

 back and sides of the body segments, but also on the sides of the abdominal legs. The occurrence 

 of such hairs in this genus is interesting from the fact that they have not yet been observed in 

 Ar(!tians, to which this moth has been referred, nor in the Noctuidw, among which it should be 

 placed, since no Arctians have when hatched smooth glanduhir hairs.' 



IV.— ON THE IXCOXGRUEXCE BETWEEN THE LARVAL AXI) ADULT CHARACTERS OF XOTODONTIAXS. 



As is well known to zoologists, from the writings of Fritz Miiller and later students, in groups 

 of animals which generally undergo a metamorphosis, two or more species of the same genus may 

 difier remarkably in respect to their early life, one species jiassing through a complicated meta- 

 morphosis while a clo.sely allied form has a direct development, hatching in the form of the adult. 

 The embryo, however, iu the latter case rapidly passes through a series of changes, constituting 

 a premature, abbreviated, or condensed metamorphosis, epitomizing the ordinary early stage of its 

 metamorphic allies. Thus the lobster differs from the other marine macruran Crustacea in having a 

 condensed metamoi'phosis before hatching from the egg, rapidly passing through a nauplius and 

 a zo(>a phase. It is so with some crabs. All the fresh- water Decapoda, notably the crayfish, have 

 no postenibryonic metamorphosis. The fact that the embryo exhibits a condensed metamorphosis 

 shows their origin from metamorphic forms. 



These are ])erhn])S the most remarkable cases of incongruence between what may be closely 

 allied genera and even species. 



Also two allied species of Gammarus may differ in toto as regards the mode of segmentation 

 of the yolk, total cleavage occurring iu one marine species {G. lociisto) and partial or peripheral 

 cleavage in two fresh-water forms {G. pulex and flnriatilis). 



Exaniples of such great divergences in larval or early life, or in the condition in which the 

 animal is hatched, iu species closelj' similar iu adult life, are not uncommon iu worms, Echiuoderms, 

 Molluscs, Crustacea, besides insects, aud the phenomenon is with little doubt due to the changed 

 conditions of the environments to which forms with such exceptional modes of develoijment have 

 been exposed. 



The principle, Then, of divergence or incongruence of larval characters iu forms whose adults 

 are closely allied has been established in the lower classes of Metazoa. The most remarkable 

 and j)uzzling case, perhaps, is that of Balanoglossus, whose Tornaria larva is so nmcli like that of 

 Echiuoderms, while the adult is a protochordate animal. 



As a matter of fact this does not affect the classification of these animals. Zoologists have 

 not thrown forms with a direct development into distinct groups where tlie adults have not 

 shown any differences; at the same time no one would unite the two species recognized as such 

 which presented no easily observed differences if one had a direct and the other a metamorphic 

 development. In the present state of our knowledge it may be well to at least provisionally 

 mark the differences between the two forms, so divergent in their early life, by giving them 

 distinct names, and thus emphasizing the fact that of the two closely allied forms one has 

 diverged from the other through having been subjected to a different set of external influences, 

 whatever such conditions may liave been. 



Systematic zoology has undergone within the last thirty years an entire change. Our present 

 systems of classification are now attempts to arrange animals in the order of their probable 

 appearance, i. e., phylogenetically, and as the subject is yet in its infancy, and our attemi)ts 

 provisional and tentative, we are obliged to give great weight to any differences in the larval 

 conditions of animals with a metamor^jhosis, because such differences were undoubteilly due to 

 differences in the environments of their parents. Indeed if it had not been owing to changes iu 

 the physical and biological environment, animals would never have risen beyond the dead level 

 of the lowest Protozoa. 



Such reflections as these and a knowledge of the mode of development of the lower classes 



of invertebrates are all-important to the students of insects, especially of the metamorphic orders, 



^ • 



'PI. XXXVII, fig. 12. Olandular liairs of Ceratoeia iricolor, a, from the second thoracic and first abdominal seg- 

 ment; h, those on the first and second abdominal legs. 



