the Classification of Lepidoptera. 351 
JuGAT^^ Comstock, Dyar. 
a. Micropterygidse)^ Hepialidse. 
Generalised Feenat^, Oomstock, Dyar. 
a. Psychid^.f 
13. Gossidse, Pyralidino., Tortricina, Tine- 
ina [Gelechiid^, Elachistidse'} , Lacos- 
omid8e,% Pterophoridse,Pyromorphidse, § 
MegalopygiddSj Eiicleidse [= the Coch- 
liopodidse (Limacodidse)'] . 
III. Classification by Hampson based on cliaracters 
derived from neuration. 
iNCOMPLETiE^ Chapman. 
1. MicropterygidsBj Hepialidse. 
2. a. Limacodidse [^Gochliopodidx], Zygxn- 
idse, Gastniidse, Megalopygidsej Psych- 
idse, Heterogynidsej Cossidse. 
yS. Sesiidse, Tineidse, Alucitidse, Ptero- 
phoridse. 
Now, it must be admitted that no one knows better 
than the authors of these papers that it is impossible to 
arrange any of the families in linear order so that their 
relationship may be shown. Yet, in spite of this 
knowledge, there is a tendency, visible here and there, 
especially in Mr. Hampson's paper, to make a somewhat 
connected list, Mr. Hampson even going so far as to 
state that he has numbered the families "from the 
bottom upwards in what seems to be the most natural 
order of arrangement.''^ Dr. Chapman very carefully 
avoids this, and only in the slightest degree does he 
even attempt to connect the families. He separates his 
Incomplete, Section 2 {vide ante), as being developed on 
four separate lines, and shows that the line of develop- 
ment of each has in its own direction reached a high 
level. For example, he states that the Zygsenidse (Incom- 
plete], Section I.) must in many respects take a high 
* Not dealt with by Dyar. 
-j- Considered as in the Frenat^ by Dyar, but not in Com 
stock's Generalised Frenat^. 
% Placed in Satur7iiid(J6 by Comstock. 
§ Belongs to or closely allied to Zygcenidce, as limited by Chap- 
man [excluding EucJiromiidce] . 
