54 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



and FulgoridiB. The wiugs of the lepidopterous \m\)a ni;iy be said to be in the nymph stage of 

 the ametabolous insects mentioned, since they are direct outgrowths from the tergites of the 

 segments from which they arise. If the wing-cases of any ]epi(h)]iterous i)ui)a, togetlier with the 

 nieso- or metathorax, are, before its larval skin is molted, removed and spread out, they bear, as 

 Spuler shows, a striking resemblance to those of a beetle, Termes, Psocus, or any Ileniipterons 

 nymiih. Tiiere are no traces in the jmpa of any of the isolated <'hitinous pieces inihe membrane 

 connecting the wings with the trunk, which are seen in the imago. If the wing of the immature 

 imago is removed from the pup.al wing case, it will be seen to ditt'er greatly in shape and venation 

 from that of the ])U|ia. The pui)al venation is ancestral and ijliylogenetic; that of the imago is 

 more specialized, showing tlie results of a long ])rocess of adaptation and modilication. Wo it is 

 with the ai>])endages; those of the maxilhe, labium, and of the legs differ greatly, as anyone has 

 observed who lias studied fresh pupa', as coinjiarcd with those from which the imago is ready to 

 emerge. Those of tiie jiupa show important dilferences; they are not simply cases, l)ut differ in 

 structure, and jwssibly represent the appendages of an ametabolous ancestor, a progenitor which 

 may have descended from the campodeiform ancestor of the class of insects. 



Tiie importan(!e of the pupa is also seen when we compare those of tlu^ genei alized Lepidoptera 

 with the more primitive generalized dipterous families Bibionidne, Cecidomyiidjc, TipulidtB, 

 Mycetophilida?, etc. The close resemblance between the orthoraphous Dipterous ])upa and Tineid 

 pupa attbrds strong evidence that the two orders are not only closely allied, but even that they 

 may have originated from a common ancestry, the loss of thoracic and of abdominal limbs and the 

 reduction of the head and its appendages of dipterous larv;e, as well as the reduction of the 

 hind wiugs, being due to modification from disuse. In the Dipterous i)U])a (Culex, etc.) the hind 

 l)airs of wings are nearly as well developed as those of lepidopterous pujta'. 



s. The imagiual features in the haustellate Lepidoi)tera will in general be found to cori'espond 

 with the pupal characters, though they are not so salient and striking as the latter after these 

 have been once observed and appreciated. In the moths (Heterocera) especially, the adaptative 

 characters have concealed the more fundamental or primitive characters. What we regard as 

 adaptative or secondary characters are the absence of vestiges of mandibles and of maxillary 

 paljji, coupled with the great development of the maxilhv themselves, the usually broad frenate 

 wings, and the difference iu shape of the two pairs, besides the specialization of the s(-ales, not 

 only of the wings, but of those forming the vestiture of the legs (in Noctuidie, etc.). 



0. What we regard as generalized or ancestral characters in the haustellate Lepidoptera are 

 those which have proved of especial service in studying the phylogeny of the order. Tliese are 

 the retention of neuropteroid characters, such as the square head, the small eyes, the vestigial 

 mandibles; in the Eriocephalidie, the retention of the lacinia and galea, the rete:i':ion of the 

 maxillary j)alpi; in the higher moths the elongated tliorax, tlie large metathorax, with separate 

 S(!Uta, the exserted large male genital armature of Micropteryx and of the Psycliicbe, the small 

 narrow wiugs of both pairs, and the trichopteriform venation of the more generalized Tineina 

 and of the Eriocephalida: (Protolejiidoptera) ; also as respects the markings of tlie wings, the 

 absence of highly colored spots, and even of bars crossing the wings. When, as in the highly 

 colored Tineids, the wings are spotted, they are often barred, this style of markings seen in Adela, 

 having been possibly handed down from or at least reminding us of certain beautifully orna- 

 mented and barred trichopterous genera. 



It will be seen, then, as we pass up from tlie Protolepidojitera to the butterflies, that there 

 lias been more or less extinction of neuro])teroid features and an increasing specialization of the 

 parts of the thorax, of the maxilhe, of theshai)e of tlu^ wings, including their scales and inai'kings 

 in general, spots succeeding bands and bars, brighter and more varied markings the dull uniform 

 hues of many micros and Lombyces. 



THE STEM FORMS OK PROGENITORS OF TIIE LEPIDOPTERA. 



It seems to us that in the discovery of two-lolx'd maxilhe in Eriocephala, and other anatomical 

 features we liave new data for discussing this subject, or at least for critising the view i)erha])s 

 quite generally held fhat the Lepidoptera have directly descended from the Trichoptera or from 

 forms more closely resembling them than other neuropteroid orders. 



