48 
ON THE CLASSIFICATION OF BIRDS. 
external structure of these birds separate them from all 
others : the legs are stout, and the hind claw sufficiently 
lengthened to betray an indication 
of scansorial habits. In this 
member, and in the shape of the 
nostrils {fig- 1 31. a magnified), we 
trace a marked analogy to the 
honeysuckers {Diceum Cuv.)-: this 
analogy is so strong, that it has 
been mistaken for an affinity, since 
we have a true Zosterops under the 
complete disguise of a suctorial 
creeper in the CeHhia borbonica of authors (6), 
(52.) Mniotilta, the fifth and last subgenus, re- 
mains to be illustrated. We have seen, in Zosterops, 
some indications of a climbing structure, and in one 
^ecies a lengthened and very slender bill; these cha- 
racters, although they by no means prepare us for the 
fiftli type of Sghrioola, will render the transition to the 
Mniotiltamria\ieil,or black and white creeper of North 
America {fig.lM. a), less abrupt than it would otherwise 
appear. This bird, in our systems, 
has shared the fate of numberless 
others, presenting a union of ap. 
parent contradictory characters. 
Although a scansorial bird, like se- 
veral others in this family, its true 
affinities unquestionably lay with 
the Sglvicolre : its bill, although 
somewhat lengthened, is notched, 
and formed on a similar model ; so 
also are its wings and tail ; its plumage likewise is totally 
devoid of those colours so universal among the true scan- 
sorial creepers, Certhiadts. But even admitting all this, 
it may be urged, in what manner has nature connected this 
bird with the Vermivorm, or worm-eating warblers ? for 
this affinity is absolutely necessary to complete the circle 
of the genus Sglvicola ? In reply to this, we must direct 
the reader’s attention to the peculiarly interesting eco- 
