30 THE BIOLOGY OF BIRDS 



other cases, it occurs in mid-winter. In reptiles what is 

 moulted is the outermost dead layer of epidermis covering 

 the scales, but in birds it is the whole feather. Thus the 

 moulting of feathers is nearer the casting of hair in mammals. 



The moulting is usually spread over a considerable 

 period, but sometimes it is so rapid that the bird is left 

 very naked. Moulting geese, ducks, and rails lose all 

 their quills at once, and are for a time unable to fly. 



The condition of things in penguins is very remarkable. 

 It has been shown by Mr. W. P. Pycraft (1907) that penguins 

 develop two successive down-plumages before assuming 

 the normal definitive feathers. Another remarkable fact 

 is that, in the penguin's moulting the feathers are not cast 

 a few at a time, but in patches. Over large areas the feathers 

 lose all direct attachment to the body, and stand out at right 

 angles or thereabouts. It may be noted that penguins are 

 antique birds, and that the distribution of the feathers (the 

 pterylosis) is more primitive than that of other Carinatae. 

 For although the penguins are flightless their affinities are 

 with the flying birds. 



Many birds moult, more or less completely, more than 

 once a year ; thus the garden warbler moults twice. There 

 is often a partial or complete spring moult, after which the 

 male birds put on special decorations for the courtship 

 season, as may be illustrated by ruff and golden plover. 

 The ptarmigan changes its plumage three times in the year ; 

 after the breeding season the plumage is predominantly 

 grey ; in the winter it is almost completely white ; in spring 

 the mottled brow^n wedding robes are put on. Mr. Pycraft 

 suggests that there may have been to begin with three or more 

 moults in the year, and that the number was reduced as the 

 feathers were perfected. The reduction in number would 

 imply a very advantageous physiological economy. 



The suggestion readily arises that the " eclipse " 

 plumage of a bird like the mallard may be correlated with 

 some seasonal change in the testes. But this is not exactly 

 confirmed by the observations of C. G. Seligmann and 

 S. G. Shattock (1914), who have looked into the matter. 



