MECHANISM OF PLUMAGE IN WATER BIRDS. 203 



of physics will remember that the surface-film of water is ap- 

 preciably tenacious, and is with difficulty broken by a fine- 

 meshed fabric. The water-repelling property of the plumage 

 of such a bird as a Duck is attributed to the grease or oil on the 

 feathers. This I find to be by no means always the case ; the 

 grease must play a very minor part indeed in the work of keep- 

 ing the feathers dry. The experiment of washing the feathers 

 of a Duck in warm soda water, and afterwards with benzine, so 

 that all trace of oil is removed, proves that the power lies in the 

 actual structure of the feathers. The parts employed are the 

 cilia and barbicels normally, the modified barbules frequently, 

 and the hamuli (when present on the contour feathers) also 

 frequently. I may be allowed to point out that this use of these 

 feather elements has not before been noticed ; and I think that 

 without their aid existence under present conditions would be 

 impossible to most birds. 



I have termed this outer mesh of barbules, and the pile of 

 cilia, the " feather film." It is hardly correct to say that when 

 a bird gets wet the water penetrates the feathers ; as a matter 

 of fact, it is the other way about — the feathers penetrate the 

 water ! As long as the tough surface-film of the water remains 

 intact, the feathers will remain dry, and the fine pile of the 

 cilia and barbicels, only to be detected by the microscope, keeps 

 the water from the coarse touch of the harder parts of the 

 feathers. 



Between this " feather film" and the skin of the bird there is 

 a thick layer of air, varying in quantity according to circum- 

 stances. In a clean and dry Black-necked Grebe I ascertained 

 it to be 4*8 cubic inches. In life this aerial envelope could have 

 been considerably increased at will. Each of the contour feathers 

 is provided with a separate apparatus of muscles, whereby it can 

 be held out at right angles or pressed close to the body. In the 

 first case the bird would appear round and fat, in the second 

 very slim, and there would be a corresponding change in the 

 extent of the air-envelope, and consequently of the buoyancy of 

 the bird. By adjusting the thickness of this layer of air be- 

 tween the " feather film " and the epidermis the bird can alter 

 its specific gravity ; and here, I think, we have the answer to 

 the riddle. 



r 2 



