40 J MY LIFE 



lies of birds and a number of the genera of other families be- 

 long to the first class, of brightly coloured birds with sexes 

 alike, and they all build in boles or make domed nests. Most 

 « i these are tropical, but the woodpeckers and kingfishers 

 are European. In the second class, however brilliant the 

 male may be, if the nest is open to view, the female is 

 always plainly coloured sometimes so much so as to be hardly 

 recognizable as the same species. This is especially the case 

 in such birds as the brilliant South American chatterers and 

 the Eastern pheasants and paradise birds. This law is of 

 especial value, as showing the exceptional need of protection 

 of female birds as well as butterflies, and the remarkable way 

 in which the colours of both classes of animals have become 

 modified in accordance with this necessity. This paper forms 

 chapter vi. of my " Natural Selection and Tropical Nature." 



5. In the great subject of the origin, use, and purport of 

 the colours of animals, there are several branches which, I 

 believe, I was the first to call special attention to. The most 

 important of these was the establishment of the class of what 

 I termed " Recognition colours," which are of importance in 

 affording means for the young to find their parents, the sexes 

 each other, and strayed individuals of returning to the group 

 or flock to which they belong. But perhaps even more im- 

 portant is the use of these special markings or colours during 

 the process of the development of new species adapted to 

 slightly different conditions, by checking intercrossing between 

 them while in process of development. It thus affords an ex- 

 planation of the almost universal rule, that closely allied species 

 differ in colour or marking even when the external structural 

 differences are exceedingly slight or quite undiscoverable. 

 The same principle also explains the general symmetry in the 

 markings of animals in a state of nature, while under domes- 

 tication it often disappears ; difference of colour or marking on 

 the two sides would render recognition difficult. This princi- 

 ple was first stated in my article on " The Colours of Animals 

 and Sexual Selection " (in " Natural Selection and Tropical 

 Nature," 1878) and more fully developed in " Darwinism." 



