ON EXTERNAL AFFINITIES. 235 
of the natural system as have been extensively demon- 
strated. 
(289.) The nature of external affinities was first in- 
timated in the following passage :—‘ I must now advert 
to another, and, in my estimation, an unquestionable 
principle of natural arrangement: this is, the direct 
union of typical groups, without the intervention of 
those which are aberrant. A vague suspicion of some- 
thing like this first occurred to me when studying the 
affinities of the Laniade, in the year 1824.* This 
property, however, belongs to very few groups, since it 
has only been detected in such as are pre-eminently 
abundant in species, and are not of a higher rank than 
families. Beyond such groups, the higher we ascend 
the more dissimilar are the typical groups in approxi- 
mating circles, until, in looking to a diagram of the ver- 
tebrated animals, not only does all appearance of affinity 
between the external or typical divisions vanish, but it 
becomes even difficult, in some instances, to trace their 
analogy. The theory of external affinities, however, 
belongs to a question so abstruse, and requires such 
nicety of investigation, that, for the present, I should 
rather, perhaps, put it as a query, than consider it as a 
demonstrated fact. At the same time, I must confess 
my utter inability to reconcile, by any other theory, the 
evident and universally acknowledged affinity between 
the Thamnophiline and the Myotherine ; between the 
Meruline and the Philomeline, and more especially 
between the typical Setophage and the typical Sylvicole. 
Unless these affinities, which I have fully detailed, can 
be disproved or explained by some other mode of rea- 
soning, it seems to be impossible to arrive at any other 
conclusion.’ ¢ 
(290.) From the above theory on external affinities 
would result another principle of natural arrangement, 
superadded to those we have already explained as belong- 
ing to natural groups. On this principle we shall not, 
* See Zool. Journ. vol. i. p. 302. 
' $ Swainson, in North. Zool. vol. ii. pref. li. 
