348 ON THE CLASSIFICATION OF BIRRS. 
find in every groupj whether large or small, these 
aberrant divisions, or forms, which represent the other 
primary types of nature. How completely this law is 
verified in the circle before us will presently appear. 
Now it is upon the analysis we shall ultimately give 
of this tribe that we wish the accuracy of our whole 
arrangement of the feathered creation to repose. All 
the details regarding their internal affinities must, of 
necessity, he included in our systematic arrangement of 
birds, intended for the second volume ; but the analo- 
gies resulting from this analysis are so highly interest- 
ing, and, at the same time, so demonstrative of the 
propositions stated in a former volume on the laws of 
natural classification*, that they must carry conviction 
to the mind of every naturalist who is at the trouble of 
investigating the facts upon which they are founded. 
(2.QO.) Analogical relations of a group are not, as some 
writers suppose, the cause, but the effect, of its natural ar- 
rangement. The clue of affinity is first sought for, and 
then, and not till then, do we seek for a uniform law of 
variation in its analogies. It would, therefore, be better, 
perhaps, had we began with giving the affinities of our 
groups in the first instance, and then have proceeded to 
show what analogical results could be drawn from them : 
the cause would then precede the effect. On the other 
hand, it may be said that, by stating the analogies in 
tile first instance, we are, in fact, stating only so many 
propositions, the correctness of which is subsequently 
to be made good by the process of analysing affinities. 
It was under this last point of view that we acted upon 
the plan we shall now continue ; that is, of stating the 
analogies in the first instance, and the affinities after- 
wards j although it is obvious that the former could 
never have been attained, without the latter had been 
first studied. In first instituting a most extensive com- 
parison of the Dentirostres with the other tribes in 
eluded in this order, we shall have occasion to notice 
many groups, whose details will be entered upon after 
• Classification of .\nh 1 nal 3 . ; 
