COLORATION 103 



is well shown in the accompanying photographs of young 

 gulls, terns, and Norfolk plovers. It is obvious that a 

 striped livery in the case of the guU or tern would be 

 the reverse of protective, while the sandy hue and narrow 

 stripes of the Norfolk plover, on the other hand, form 

 a most effective disguise, the stripes bearing a close 

 resemblance to the sparse blades of grass or other vegetation 

 such as is met with on sandy wastes. 



It may be objected that the persistence of stripes in 

 the Cereopsis goose — when all other Anatidae have lost 

 them — and in the young osprey, when all other Accipitres 

 are whole-coloured, and commonly white — and the black 

 mantle of the young coot, which may be reared in a nest 

 within a yard of that of the young grebe, which is striped, 

 throws a strong element of suspicion on the " protective 

 resemblance " theory. The reply of those who have 

 studied these nestlings in the field would be that whatever 

 may be the ultimate interpretation of these puzzUng 

 exceptions, the reality of the protective value of this 

 coloration in the case of grebes, gulls, terns, plovers, and 

 so on, is incontrovertible. 



Before we can cite the exceptions as disproving the 

 rule we must gather more facts as to the life-history of 

 the birds which present them. One fact must not be lost 

 sight of in this connection, and this is that ancestral 

 characters may persist long after they have ceased to 

 fulfil their original function, if by their presence they 

 do not imperil the well-being of the animal which retains 

 them. These persistent and apparently non-significant 

 colours indicate, surely, either that they are useful or 

 that the coloration in such cases is a matter of no 

 importance, so that the ancestral livery has not been 

 eliminated as harmful by natural selection. 



