ON THE CLASSIFICATION OP BIRDS. 
dication of this ornament is seen in those birds where 
the head is very full of feathers, which are somewhat 
longer than those of the nape and surrounding parts, 
without assuming any particular shape; the appa- 
rently large head of very many birds is caused by this 
superabundance of feathers, rather than from the size 
of the skull. The head of the Tamatias, or puff- 
birds of Brazil, when they are quietly seated upon a 
branch, appears to be nearly double the natural size, 
from its thick clothing of feathers, which are then 
somewhat raised*; but if the bird is suddeidy disturbed, 
instead of erecting these feathers still more, as the 
woodpeckers do, the puff-bird immediately brings them 
flat upon the skull ; and its head, although at all times 
large, seems suddenly reduced to half its size.f The 
thick-heads of New Holland {Pachnnephula Sw.), 
which represent the last mentioned group in their own 
circle, have the feathers equally thick, but we know 
nothing of their manners. We can hardly call this 
formation a crest, seeing that it assumes no form inde- 
pendent of that common to the head, and it is certainly 
no ornament. There can be no doubt, however, that 
it is connected, in some way, with the economy of the 
birds possessing it, otherwise we may feel assured it 
would not be possessed only by certain races. Incipient 
crests, in the same manner, are possessed by many birds. 
which are generally de- 
scribed, even by system- 
atic writers, ashaving none. 
We are not aware that any 
of the describers of Bri- 
# tish birds have mentioned 
the crest of the male chaf- 
flnch+ {fiy. 12.), which 
is always erected when the 
bird is disturbed, and retained in that position even after 
