130 
SEXUAL selection: birds. 
Part II. 
yarious hideous deformities — deep scars on the face 
with the flesh raised into protuberances, the septum 
of the nose pierced by sticks or bones, holes in the ears 
and lips stretched widely open — are all admired as 
ornamental. 
Whether or not .unimportant differences between the 
sexes, such as those just specified, have been preserved 
through sexual selection, these differences, as well as 
all others, must primarily depend on the laws of varia- 
tion. On the principle of correlated development, the 
plumage often varies on different parts of the body, or 
over the whole body, in the same manner. We see 
this well illustrated in certain breeds of the fowl. In 
all the breeds the feathers on the neck and loins of 
the males are elongated, and are called hackles; now 
when both sexes acquire a top-knot, which is a new 
character in the genus, the feathers on the head of the 
male become hackle-shaped, evidently on the principle 
of correlation ; whilst those on the head of the female 
are of the ordinary shape. The colour also of the 
hackles forming the top-knot of the male, is often cor- 
related with that of the hackles on the neck and loins, 
as may be seen by comparing these feathers in the 
Golden and Silver-spangled Polish, the Houdans, and 
Creve-coeur breeds. In some natural species we may 
observe exactly the same correlation in the colours of 
these same feathers, as in the males of the splendid 
Golden and Amherst pheasants. 
The structure of each individual feather generally 
causes any change in its colouring to be symmetrical; 
we see this in the various laced, spangled, and pen- 
cilled breeds of the fowl ; and on the principle of 
correlation the feathers over the whole body are often 
modified in the same manner. We are thus enabled 
without much trouble to rear breeds with their plum- 
