272 ¥IRST PRINCIPLES OF NATURAL CLASSIFICATION. 
doubt, be easy to divide the family of warblers in twenty 
different ways, if mere division were the object; and thus 
to increase or diminish the number of the groups: but 
to do this in more ways than one, so that each division 
shall be a circular group, is utterly impossible ; and it 
therefore follows, that these groups, and their relative 
value, exist as truly and as absolutely in nature, and are 
as capable of being defined, as quadrupeds are from 
birds, or fishes from insects. It will be our object, at 
a more convenient season, to exhibit an analysis of the 
second column. This analysis, in short, has been made: 
and as the results have been procured by the same 
means, we could not withhold from our readers so in- 
teresting a coincidence of the definite number and rank 
of natural groups, taken from two different classes ; al- 
though the truth of one must at present be assumed. 
-"(332.) Although we have not yet been able to detect 
any circular groups below the rank of sub-genera, the 
species composing these little assemblages, when they 
are numerous, always present us with representations of 
those primary types of form already defined, and to 
which nature so tenaciously adheres. We see an ex- 
ample of this in the bearded titmouse, which is the 
natatomtal or aquatic type of a sub-genus ; while Poda- 
lirius Machaon, in like manner, is an aquatic or fissi- 
rostral type. The student must not, however, look to 
these two series with any expectation of tracing trans- 
verse analogies: for although the two orders of Jnses- 
sores and Lepidoptera really happen to represent each 
other, and are typical orders, the other groups are of 
different denominations ; and therefore cannot, as we 
have already explained, be analogous. 
(333.) In further proof of the universality of these 
groups, both as regards their number and their rank, we 
shall now produce two other series: one of which is 
selected from the class of Mollusca, for the purpose of 
ascertaining the rank of that lovely shell, the Harpula 
Vexvillum Swains. ; and the other from the sub-kingdem 
