416 ENTOMOLOGICAL SOCIETY 



Food plant Viburnum prunifolium. The larvae absolutely re 

 fused V. lentago and V. dentatum. 



Found along Pimmit Run in Virginia and near Hyattsville, 

 Maryland. 



The life history of this species is very interesting in connection 

 with Tutt's classification of Lepidoptera ( J. W. Tutt, A Natural 

 History of British Lepidoptera, 1899, p. 113). Tutt divides the 

 order into three main lines of descent, all of which show both 

 the generalized and specialized characters of the several stages, 

 except the egg. The egg Tutt regards as absolute. He quotes 

 approvingly Chapman's suggestion, who " thinks it is probable 

 that the various forms of the upright egg had a common origin, 

 though very low down." While not disputing Chapman's valu 

 able generalizations, it seems scarcely more probable that the 

 upright form of egg should have been developed but once in the 

 evolution of the Lepidoptera than that the obtect pupa, or the 

 four-branched median vein, or the post-spiracular position of 

 tubercle iv should have been so developed. All of these Tutt 

 freely derives more than once in his genealogical tree, but not so 

 with the egg. Now, however, with regard to Callidapteryx. We 

 have here a moth that has been classified with the Geometridse 

 and bears so close a relation to them that Hulst still lists it in a 

 subfamily. Hampson separates it as an adjacent family, but only 

 on a character of venation of seemingly secondary importance. 

 Judging from the moth, this insect would fall in Tutt's " Geo- 

 metro-Eriocranid stirps." The larva, as I have shown, possesses 

 some peculiar generalized characters. I can hardly tell where 

 Tutt would place it on his tree, but presumably in the stirps just 

 mentioned. The egg, on the contrary, is upright and ribbed and 

 would come high up in the " Noctuo-Hepialid stirps." I think 

 that the occurrence of this generalized form connecting the Geo- 

 metrids and Noctuids shows that Tutt's " stirps" are not entirely 

 natural ones ; his genealogy of the Lepidoptera has been too 

 much influenced by Chapman's brilliant results on the egg. 'If 

 these egg characters be given a less important possition, similar 

 to that of the other characters, it is possible to recast Tutt's tree 

 so as to bring the larvse together more in harmony with the ideas 

 that I have expressed. I add a chart, showing the derivation of 

 the families of the Bombyces which is the same as that recently 

 published (Proc. Bost. Soc. Nat. Hist., 27, 146) but modified to 

 exclude the contradiction in egg characters to which Tutt rightly 

 objects. 



The tree is supposed to start from some generalized Tineid- 

 like form. The Nolidas branch off first, being really of separate 

 origin from the Bombyces and possessing Tineid characters, yet 

 so highly evolved as imagines as to parallel the Arctiid structure. 

 The Geometridae, Drepanidae, and Epiplemidae form a closely 



