360 Mr. J. W. Tutt on recent papers on 
new and more natural reconstruction into families^ which 
I hope will be done before long. As far as I am able to 
judge, however, the Tineidse represent the ramifications 
of one branch of the Lepidoptera, some families gene- 
ralised, others highly specialised, and not a heterogeneous 
collection of famihes sprung from various parts of the 
Lepidopterous tree as the old family Bombyces did. 
*^ The same remark applies to the Ehopalocera. I left 
them alone because I had not studied them; but here 
I am doubtful if we have not four different stocks : the 
Hesperidse arising from the Castniidse ; the Erycinidse and 
the Lycsenid.se from near the Callidulidse ; the Papilion- 
idse, as also the Pieridse and Nymphalidas from the 
Zygseno-Cossid stock ; but these suggestions are not 
based on any very careful examination. 
" The Lasiocampidse, Endromidse, and Arhelidse are out 
of place in my artificial key, because in such a key it is 
always practically impossible to get quite a natural order, 
and I have not yet found a character common to them, 
and not found in other families, which would place them 
together in their natural position at the bottom of the 
Obtect^. The Lasiocampidse and Endromidx I believe 
to have developed from the Limacodidsej the Arhelidse from 
the Cossidse. 
There is a point in your paper that I entirely dis- 
agree with, the larva is not an embryo, being subject to 
the struggle for existence and to modification in relation 
to environment, and arguments based on superficial larval 
characters, such as Mr. Dyar's setiferous tubercles, are 
in no sense on a par with arguments from embryology; 
and if the results he had obtained had been entirely 
negative, it would have shown that the character he ex- 
amined was faulty and would not have invalidated 
the results obtained from the neuration of the imago 
which is not changed by external circumstances, except 
to some degree when the shape of the wing is much 
modified. 
" Dr. Chapman^s pupal characters are not so liable to 
this objection, especially the mode of dehiscence, the 
pupa itself being more of the nature of an embryo, and 
the methods of emerging from the pupa are less liable 
to the action of natural selection; but I believe these 
characters of his IncomplettE, as well as the emergence 
of the pupae from the cocoon, primarily to have relation 
