NESTLING FEATHERS OF THE MALLARD. 613 



protoptile of a twenty-six clays' duck embryo artificially removed 

 from a filament (like the long filaments in PI, III. fig. 9) is 

 represented in PI. III. tig. 10. 



The wing-quill protoptiles are complete feathers : i. e., each 

 consists of a calamus, a shaft, and an aftershaft. The calamus, 

 though only 3 or 4 mm. in length, is well formed, contains 

 several " cones," and remains intact imtil the protoptile is shed — 

 in Penguins the part of the epidermic tube which represents a 

 " calamus " usually soon splits, with the result that the individual 

 protoptile barbs are directly continuous with mesoptile barbs 

 (text-tig. 14). Though the growth of the shaft and aftershaft is 

 arrested before hatching, the calamus may continue to grow after 

 hatching. 



The shaft is made up of a rhachis, continuous with the outer 

 segment of the calamus, and usually of seven or eight pairs of 

 barbs, which vary in length and in the nuitaber of barbules they 

 possess. The two terminal barbs (which by uniting formed the 

 distal part of the rhachis) end in long slender processes destitute 

 of barbules. The aftershaft (PI. III. fig. 11), continuous with the 

 inner segment of the calamus, usually consists of eight barbs 

 bearing barbules. Sometimes the four mesial barbs of the after- 

 shafts iTuite to form a short rhachis. The bai'bules of the shaft 

 are spirally twisted at their origin (text-fig. 1), but the aftershaft- 

 barbules only curve slightly on leaving the barb. It has frequently 

 been stated that aftershaft-barbules of true as well as nestling 

 feathers have no cilia, but cilia are invariably present on the 

 barbules of the protoptile aftershaft of Mallard ducklings. Part 

 of a barbule with cilia is given in text-fig. 2. 



The Aving-quill protoptiles may all be present at the end of the 

 sixth week (PI. III. fig. 12), but some of them have usually been 

 shed before the end of the seventh week. Though Mallards 

 begin to fly during the ninth week, the development of some of 

 the wing-quills is only completed during the tenth week. 



(2) The Wing-Quill Mesoptiles. 



Up to 1906, as already mentioned, it had not occui-red to 

 ornithologists that true feathers (pennje) might be preceded by 

 two generations of nestling feathers (prepennse), and mesoptiles 

 have apparently not yet been recorded in connection with wing- 

 quills. 



Early in the century two National Antarctic Expeditions were 

 fortunate enough to collect material which made possible a fairly 

 exhaustive study of the development and history of the nestling 

 feathers of Penguins. The penguin material brought home by 

 the English National Antarctic Expedition was reported on by 

 Mr. Pycraft ; that collected by the Scottish National Antarctic 

 Expedition by Dr. Eagle Clarke. In a paper published in 1906 

 in ' The Ibis,' Dr. Clarke announced that he had discovered two 

 coats of nestling feathers in the Ringed and Gentoo Penguins, 



