1883.] Note on the Classification of Moths. 497 
having exclusive characters, and hence the effort of late to break 
up the Bombyces into distinct families. Thus disassociated, the 
general character of “form” as laid down by Agassiz neglected, 
the next step is to force them into unnatural positions, to displace 
the Hepialinæ, and rank them with the Castnians, or to produce 
such an arrangement as Von Heineman proposes. The “ sub- 
families” of the Noctuidze are, on the other hand, vague, and 
they shade into each other, in such a way, that it becomes a mat- 
ter of choice where we assign the limit. The difficulty lies 
mostly with the Noctuidæ. Two groups, Bombyciz and Noctuo- 
phalænidi, possess exclusive characters which have led to their 
being treated as distinct families by the German entomologists. 
In form these two small groups are decidedly noctuidous. The 
one closely resembles the higher, the other the Catocaline group 
of the Noctuidae. Certain general assemblages, such as the Bomby- 
coidea, Hadeninz, etc., are, in a general way, admitted by Lederer, 
denies them a scientific basis, and discards them in the 
body of his work. Excluding the two groups above noted, he has 
G division of the family at all higher than a generic one, The 
‘SMe is true of Geometridz, although here the divisions show 
More character, In the Pyralidz, however, the sub-family groups 
are again strong. The existence of these sub-family groups are, 
) Perhaps, dependent on the general question of the development 
“the suborder. The intermediate groups have fallen out and 
_ Sasedto be perpetuated, 
ae There '$a certain parallelism between the Sphingidz and Noc- 
: the ie from their usually naked larve, the hairy larve of 
om. t Noctuids having a Bombycid analogy which Butler 
_.. » Fegards as a real affinity. 
: : Pe rope “Mural classification of the Noctuidæ, seems to me 
i eo in my “ Check ‘List of 1876,” where I recognize 
Packard Soma groups, the Noctuinz and Catocaline of Dr. 
te in the Deltoides of Latreille; considering the other 
Value. e above mentioned, as having a higher than sub-family 
TR difficulty lies in the existence of groups intermediate 
ea ily and Sub-family divisions as hitherto assumed. In 
Wa ‘Characters, the different values between these groups 
Before ect: Our terminology will have to be extended ; but 
ees undertaken, m isi ith 
a » More precision must be attempted wi 
MS we have on hand, 
