1898] THE NEURATION OF RHOPALOCERA 395 
a four-branched cubitus in the forewings, the generalized cubitus- 
anal connection, and the two anal nervures of the forewings being 
retained, likewise the subcostal-radius connection of the hindwings. 
One anal nervure only, however, is retained in the hindwings; that 
is to say, one of those present in Nymphalid-Pierids has been lost. 
My conclusion from a study of the neuration is that the distinct 
separation of Papilionids in phylogeny from the other Rhopalocera 
eannot hold good, and in this I join issue with Dr Grote. 
Whether I have clearly demonstrated this in the above, I cannot 
say; but the facts are distinctly in evidence that, however widely 
the specialized forms differ, the generalized forms: of neuration 
indicate a natural sequence in modification—the evolution of the 
Papilionid form from a primaeval form such as is illustrated by 
Cossids; and of this evolution the generalized Hesperids (1), 
Nymphalid-Pierid (2), and finally generalized Papilionids represent 
the surviving links in the continuity of specialization, a primary 
modification of neuration antecedent to and quite apart from the 
special modifications peculiar to the several groups. 
I need only refer to Dr Chapman’s paper, “ Butterfly Pupae” 
(Entom. Record and Journal of Variation, vol. vi. pp. 106 and 125), 
to support my conclusions: “The lowest (ze. most ancient) forms 
in all the families are really very close together.”——“ Pierid and 
Nymphalid started together, shortly afterwards separating.” Nym- 
phalid pupae are capable of lateral movement only, as are Pierids, 
but have lost the ‘girth’ which is characteristic of the Pierid and 
Papilionid method of pupation; Nymphalids, however, retain the 
‘double nosehorn’ pupal structure of Papilionids (7c. a generalized 
pupal character), Pierids having lost this particular pupal character. 
The ‘ double nosehorn’ is surely corroborative of my conclusion that 
the features A, B, C of Anosia are homologous with the Papilionid 
cubitus-anal connection, short anal nervure in forewings, and sub- 
costal-radius connection in hindwings, which establish affinity be- 
tween the Nymphalid-Pierid and Papilionid groups. 
If Papilionids be separated from Nymphalid-Pierids in the man- 
ner proposed by Grote, then the neuration of Papilionids must be 
regarded as isolated, highly specialized, and without surviving forms 
connecting it with primaeval neuration from which it has been 
derived. On the other hand, the “primary modification of 
neuration” among primitive Rhopalocera shows the formation of 
the three-branched ecubitus in forewings, with two anal nervures 
in hindwings; then follows the formation of the four-branched 
cubitus, with loss of one anal nervure in hindwings. This is 
more in accord, I believe, with the evidence of development, and 
supplies the connecting links between the primaeval neuration and 
that of the highly specialized Papilionid. A. QUAIL, 
PALMERSTON NortH, NEW ZEALAND. 
