278 STUDY OF NATURAL HISTORY. 
must admit that his views on its relative affinities are 
any thing but artificial. Metamorphosis, in fact, is 
really one of the primary distinctions of the typical 
Annulosa, but it is not the only one; so, also, 
is the structure of the mouth. Yet neither of these, 
by themselves, will completely designate the typical 
groups. We know, by experience, that every 
peculiarity or variation in metamorphosis is almost 
always accompanied by external differences in 
structure, permanent in themselves, and always 
within reach of observation. Why, therefore, should 
we designate our groups by characters which are 
evanescent, when the same object can be attained 
by using others that are permanent? How is a 
student, for example, to discover the natural tribe 
to which any particular beetle belongs, and of 
whose metamorphosis he is entirely ignorant, if the 
tribes are to be characterised alone by their meta- 
morphosis—that is, by the form of their caterpillar 
and chrysalis? The thing is manifestly impossible. 
But the evils of assuming this theory as infallible 
do not stop here. One of its most able and in- 
telligent advocates has made metamorphosis the 
basis of his arrangement of the Lepidoptera; so 
that, if this plan be generally adopted, we shall never 
feel certain on the natural affinities of an insect, 
until we have studied its larva and pupa. For our 
own part, we must confess that we have the greatest 
objection to such characters; and we think that 
it is the duty of every naturalist to simplify the ac- 
quisition of science, by choosing such characters for 
his groups, as can be easily understood, and at all 
times verified. Now this cannot, of course, be said 
