March 1897.] GROTE I CLASSIFICATION OF THE SaTURNIIDES. 45 
the rest. And this proves the value, that the character does not 
fail. * 
The adverse statement fails, when I show, that in the larval special¬ 
ization (the diminution of the tubercles and armature), the antennal 
structure (the attainment of the equally lengthy pectinations), the nega¬ 
tion and the complexity in the attachment of the cocoon, a consonant 
direction is held and a perfectional advance throughout the Saturniidce 
(including Hei?iileucd). Dr. Dyar’s statement that I have transposed the 
position accorded by him to Hemileuca and Aglia is strictly correct and, 
as I try to show here, entirely defensible. The former, Dr. Dyar would 
place with the Automeris group on account of the stinging spines. But 
I prefer to consider the eversible glands and stinging spines of the cater¬ 
pillar as here characters of convergence. Their presence is explainable 
by the consideration that both Hemileuca and Automeris have probably 
arisen or diverged from a common point nearer the basis of the phyl- 
lum. It is easier to see that the stinging spines are a subordinate char¬ 
acter when we find them again in unrelated groups: e. g. Apodidce. It 
is not possible for me to “ suppose that vein IV 2 has moved towards 
[V 1 in Hemileuca separately from the type of Attacus and Saturnia 
where this process is congenital.” Since I show that the type is fully 
attained in Hemileuca , it is plainly already congenital in the Hemileu- 
tinee. The real morphological value of this “ movement” is strangely 
underrated by Dr. Dyar. In reality it is profound. It amounts to a 
reorganization of the wing through the action of the Radius upon an¬ 
other pattern. Ip a paper subsequently read by me at the Frankfort 
meeting, I have tried to trace the process by which the lower and more 
generalized Agliid wing has passed into the higher, more specialized 
Saturniid type. The difference, as we now find it, is, relatively speak- 
mg, primary, palingenetic, not adaptory and secondary, as appears to me 
the change of the armature into stinging spines. 
With reference to Aglia, which I believe to be a specialized and 
very much isolated type, I regard it as having left the main Agliid 
stem before the devolution of Citheroma as we now find this group. 
The loss of the pair of anal tubercles is to be set down solely to the 
Citheroniince. I do not derive Algia from Citheronia , but from the 
stem before Citheronia. Dr. Dyar charges me with entertaining more 
* Since my paper went to press, the Roemer Museum has received additional 
material of South American Saturniides in all stages. In a paper read September 
2 3 ( 1 > a t the Frankfort meeting, I show that in all the new material the characters 
pointed out by me hold good and sustain my general classification. 
