158 Journal New York Entomological Society. [Voi. v. 



or Limnadidas, which are distinctly less specialized than the Nymphalidae 

 proper. So that we see that the statement of the Editor of the Phila- 

 delphia "Check List," that, in his "opinion," the Nymphalidse are 

 " correctly placed at the head of the Rhopalocera " is not derived from 

 what this writer elsewhere calls "scientific knowledge" or "science," 

 but is plainly the result of an effort to get into good company. It is 

 characteristic also of this sort of "opinion," that when we turn to the 

 List itself we find it to "head" with the Limnadidae, the most gener- 

 alized of the four-footed Butterflies. The success of the Nymphalid 

 branch in attaining a variety of forms and a vast array of species 

 has been great, and this tends to our believing it to he so dominating. 

 It is, however, lateral, not on the main line. In the accompanying 

 diagram the opening of the cell has led me even to give the higher 

 groups perhaps too exalted a position, but this is a minor point. The 

 connection of the Charaxini, a foreign group, with the main stem of the 

 Nymphalidce is problematical. I have commented on its position else- 

 where, and it must be brought into place when the tropical butterflies 

 are studied upon the basis here set forth. 



4. It may be further assumed, that, in former periods of time, the 

 grouping was laxer than to-day, and that the families we now are able 

 to separate were once interconnected by forms which have dropped out. 

 At that time the four or brush-footed butterflies may have been more 

 nearly connected with the six-footed stem. From small and specialized 

 groups we cannot expect the birth of new features, but from large and 

 spreading assemblages, presenting a wide range of character. That 

 such a state of aff"airs existed in the Whites, we have the testimony of 

 Leptidia to prove. This butterfly appears now as an isolated survivor 

 of what was probably a large group of Pieridse. The abyss separating 

 Leptidia from the Pierinae is profound and I am informed that even 

 more important deviations still exist in the family. The Pierids may 

 then well represent the matrix from which the four-footed type pro- 

 ceeded. 



5. Boisduval's groups of Suspensi, Succincti, Involuti, based on the 

 fashion of fastening the chrysalis, have no existence as phylogenetic as- 

 semblages, hence are improperly used in this manner by Mr. Scudder. 

 The Papilionid, Pierid and Lycjenid Succincti have clearly reached the 

 habit independently. It is a fallacy to believe, with Mr. Scudder, that 

 there is a regular progression from the cocoon of the moths to a total ab- 

 sence of the use of silk. Instances are not rare where the generalized 

 forms spin little or no silk and the specialized forms on the same phyloge- 



