NATURAL SELECTION 



In the extensive families of the warblers (Sylviadae), 

 thrushes (Turdidae), flycatchers (Muscicapidse), and shrikes 

 (Laniadee), a considerable proportion of the species are beauti- 

 fully marked with gay and conspicuous tints, but in every 

 case the females are less gay, and are most frequently of the 

 very plainest and least conspicuous hues. Now, throughout 

 the whole of these families the nest is open, and I am not aware of 

 a single instance in which any one of these birds builds a 

 domed nest, or places it in a hole of a tree, or under ground, or in 

 any place where it is effectually concealed. 



In considering the question we are now investigating, it is 

 not necessary to take into account the larger and more power- 

 ful birds, because they seldom depend much on concealment 

 to secure their safety. In the raptorial birds bright colours 

 are as a rule absent ; and their structure and habits are such 

 as not to require any special protection for the female. The 

 larger waders are sometimes very brightly coloured in both 

 sexes ; but they are probably little subject to the attacks of 

 enemies, since the scarlet ibis, the most conspicuous of birds, 

 exist in immense quantities in South America. In game birds 

 and water-fowl, however, the females are often very plainly 

 coloured, when the males are adorned with brilliant hues ; 

 and the abnormal family of the Megapodidee offers us the in- 

 teresting fact of an identity in the colours of the sexes (which 

 in Megacephalon and Talegalla are somewhat conspicuous), in 

 conjunction with the habit of not sitting on the eggs at all. 



What the Fads Teach us 



Taking the whole body of evidence here brought forward, 

 embracing as it does almost every group of bright-coloured 

 birds, it will, I think, be admitted that the relation between 

 the two series of facts in the colouring and nidification of 

 birds has been sufficiently established. There are, it is true, 

 a few apparent and some real exceptions, which I shall con- 

 sider presently ; but they are too few and unimportant to 

 weigh much against the mass of evidence on the other side, 

 and may for the present be neglected. Let us then consider 

 what we are to do with this unexpected set of correspondences 

 between groups of phenomena which, at first sight, appear so 

 disconnected. Do they fall in with any other groups of 



