724 The Colors of Animals and Plants. [ December, 
which escaped to produce the next generation were those which 
would produce the more highly colored butterflies, it is difficult 
to perceive how the slight preponderance of color sometimes 
selected by the females should not be wholly neutralized by the 
extremely rigid selection for other qualities to which the off- 
spring in every stage are exposed. The only way in which we 
can account for the observed facts is by the supposition that 
color and ornament are strictly correlated with health, vigor, and 
general fitness to survive. We have shown that there is reason 
to believe that this is the case, and, if so, voluntary sexual selec- 
tion becomes as unnecessary as it would certainly be ineffective. 
There is one other very curious case of sexual coloring among 
birds: that, namely, in which the female is decidedly brighter 
or more strongly marked than the male, as in the fighting quails 
(Turniz), painted’ snipe (Rhynchea), two species of phalarope 
(Phalaropus),and the common cassowary ( Casuarius galeatus). 
In all these cases, it is known that the males take charge of and in- 
eubate the eggs, while the females are almost always larger and 
more pugnacious. In my Theory of Birds’ Nests? I imputed this 
difference of color to the greater need for protection by the male 
bird while incubating, to which Mr. Darwin has objected that the 
difference is not sufficient, and is not always so distributed as to be 
most effective for this purpose; and he believes that it is due to 
reversed sexual selection, that is, to the female taking the usual 
réle of the male, and being chosen for her brighter tints. We 
have already seen reason for rejecting this latter theory in every 
case, and I also admit that my theory of protection is, in this case, : 
only partially, if at all, applicable. But the general theory of in- 
tensity of color being due to general vital energy is quite appli- 3 
cable ; and the fact that the superiority of the female in this re- 
spect is quite exceptional, and is therefore probably not of very 
ancient date in any one case, will account for the difference of 
color thus produced being always comparatively slight. ae 
Theory of Typical Colors.— The remaining kinds of animal 
colors — those which can neither be classed as protective, warn- 
ing, nor sexual — are for the most part readily explained on the 
general principles of the development of color which we have «> 
now laid down. It is a most suggestive fact that, in cases where 
color is required only as a warning, as among the uneatable cat- 
erpillars, we find, not one or two glaring tints only, but every 
kind of color disposed in elegant patterns, and exhibiting almost 
1 Natural Selection, page 251. a, 
