the Classification of Lepidoptera. 
345 
^' frenulum form_, to a large extent, the basis of Pro- 
fessor ComstocVs classification; his JuGATiE containing 
the Hepialidse and Micropterygidse, whilst his Frenat^k 
are divided roughly into " Greneralised Frenulum Con- 
servers/^ "Specialised Frenulum Conservers/' and 
Frenulum Losers/^ 
It will be remembered that Dr. T. A. Chapman, from a 
study of the characters presented by the earlier stages 
(especially by the pup^) of the Lepidoptera-Heterocera, 
divided them into two groups, Obtect^ and Incomplete', 
and placed the Micropterygidse, Cochliopodidse, and 
Hepialidse among the families which he considered to be 
at the bottom of the scale of development of the Lepido- 
ptera ; whilst it has long been known that the Hepialidse 
and Micropterygidse differed much from the other families 
of Lepidoptera, inasmuch as they possessed twelve veins 
in the hindwings, no other family having more than 
eight. 
Since the publication of Professor Comstock^s paper, 
Mr. Harrison G. Dyar, S.B., has worked out a general 
scheme of classification,* based on the arrangement and 
external structure of the setiferous tubercles of the 
larvae of the Lepidoptera, in order " to see how a classi- 
fication, based on these structures, would compare with 
this new classification.^^ After pointing out that the 
special development of these tubercles is largely for the 
purpose of defence, and that, consequently, a classification 
based on larval tubercles might be expected to differ in 
important respects from one based on the adult insect, 
Mr. Dyar very rightly insists that the fact that his system 
does not differ in any important particulars ''^for the 
major groups shows that Professor Comstock^s classifi- 
cation is the nearest to a natural one that we have yet 
had.^^ 
Owing to the loss of the frenulum in certain 
Frenat^/' Prof Com stock states that it is necessary to 
make use of some other character or characters " 
which are acknowledged " by systematists as recogni- 
tion characters.^' He then falls back upon neuration, 
a line which is worked out somewhat at length by 
Mr. Hampson in the paper referred to above. 
* " A Classification of Lepidopterous Larva?," Annals New York 
Acad. Sci., viii., p. 18. 
