440 [December 1898 
CORRESPON DENCE 
THE CLASSIFICATION OF BUTTERFLIES 
In brief reply to your note on page 298, vol. xiii., I would state that the parity of 
specialisation in the feelers of the Papilionidae and Nymphalidae does not necessarily 
authorise the association of the groups. I have not separated the Papilionides (Parnas- 
siidae and Papilionidae) upon any characters of specialisation, but upon a fundamental 
divergence in the structure of the wings, such a divergence as indicates, in my opinion, 
a distinct phylogenetic line and is plainly available for taxonomic purposes. The gist 
of my article in Nalwral Science and of that in the Proceedings of the American Philoso- 
phical Society is, that parallel specialisations in any one organ do not authorise associa- 
tion, since they have been independently acquired. Rank is a relative conception. 
I do not claim that the Papilionides should outrank all the other butterflies, but I do 
claim that they cannot be logically interpolated between any of the groups of the other 
butterflies, a mode of classification as old as Roesel, revived by Bates, Distant, Scudder, 
Reuter, and others. This interpolation cannot take place because the Papilionides 
possess, in vein ix. of primaries, a morphological character unshared by all the other 
butterflies, which again have in common a loop-like appendix to vein vii. at base 
(vein viii.). This appendix fades out by disintegration in certain minor groups. It is 
also shared by the more specialised and larger groups of the moths. The butterflies are 
therefore probably dichotomous, and I expect the phylogeny of the Papilionides is to be 
sought for in the Tineides. The two groups of diurnals have attained their character as 
day fliers by different routes. 
My criticism of Dr Jordan’s investigations is, that, valuable as they are as a con- 
tribution to the morphology of the feelers, they are valueless for taxonomy and phylo- 
geny. I do not regard the specialisations of the veining, which takes a parallel direction 
so far as concerns the radius and its branches in the Lycaenidae and Pieridae, as any 
warrant for the association of these families. And I must naturally object to having my 
classification of the butterflies into two major groups, viz., Papilionides and Hesperiades, 
criticised upon the basis of coincidences in specialisation of certain organs common to 
both groups. The Papilionides possess an important part of an organ which the other 
butterflies want. Specialisations come in afterwards, as when I show that the Parnas- 
sians are more specialised than the Swallowtails, and that therefore we should com- 
mence our lists with Apollo and kindred butterflies. A. RADCLIFFE GROTE. 
RoreMER MusEum, HILpESHEIM, 7th November 1898. 
IMPORTANT NOTICE 
After this number NATURAL SCIENCE will be published by 
Mr YOUNG J. PENTLAND, 
11 Treviot PLacrt, EDINBURGH, 
AND 
38 WEST SMITHFIELD, LONDON, E.C. 
Communications for Editor and Publisher should be sent to 11 Teviot 
Place, Edinburgh. 
London parcels may be sent to 38 West Smithfield, 
