60 ON THE OIjASSIFICATION OF BIRDS. 
of the group, Setophaga ruticilla (/, 7t), or the Ame- 
rican redstart of Wilson, which every author, who has 
mentioned the bird, admits to be a flycatcher. 
(68.) Such, tlien, is the scries by which Sylvicola 
and Setophaga are united j and it has been marked out 
by one who had neither a system to propound, nor a 
theory to support. It is, indeed, most interesting to find 
that this thread of nature has been followed up with 
such admirable precision by one who has been aptly 
termed a field naturalist : for ourselves, we cau only add, 
tliat every one of these changes in the habits of the spe- 
cies here enumerated, is marked by a corresponding 
difference in some part of their external structure. 
( 69 .) The connection of Setophaga with Sylvicola 
being thus established, it remains to be considered if the 
first of these groups holds that rank in the circle of the 
PaWaniB which we have here assigned to it — that is, of 
a genus ; — or whether it is in fact but a subgenus of 
Sylvicola, really occupying the station we have given 
to Dumecola. The solution of this question is exceed- 
ingly difficult; and, although it has occupied our attention 
for many years, we by no means feel that conviction 
of the accuracy of our present decision, which we do on 
many other analogous instances. This we unequivo- 
cally mention, that those ornithologists who may here- 
after have time and opportunity for going over the same 
ground may not attach undue weight to our opinions. 
We consider, then, that this is a perfectly analogous case 
witli the affinity between Tlmmiiophilus and Myothera, 
botli of which are external genera in two different circles, 
but which do not immediately follow each other in the 
same circle. Both, in short, seem to be instances of the 
strongest external affinity, or of that relation, which in 
larger, and therefore more dissimilar, groups would be 
termed only an analogy. We are induced to form this 
conclusion regarding the two groups immediately before 
us, not so much from the perfect analogy which the case 
bears to others, but because Sylvicola appears to be al- 
most as closely connected with Dumecola as it is to Seta- 
