VI FORM AND FUNCTION I51 
pulp, as it is called, is the same in the nestling down 
and the more lasting and stronger formation that 
follows it. The change, therefore, bears no resem- 
blance to the shedding of milk teeth and their re- 
placement by permanent ones. The early tooth is 
driven out by the later one ; the two are not in any 
way connected. 
After the first moult, the feathers develop without 
any nestling “downs” as precursors. Otherwise the 
process is not, in essential points, different. The cells of 
Fic. 41.—(After Gadow). Showing development of feather. 
as, cells forming after shaft ; B, cells forming barbs; ms, cells forming main shaft ; 
sH, horny sheath surrounding whole feather. 
the papilla, or rather of the epidermis over it, arrange 
themselves starwise. Two of the columns of cells 
which cause this starlike formation grow broader and 
longer than the rest, and go to make the rachis of the 
feather. Two on the opposite side form a secondary 
shaft, of which, as I have said, most feathers retain 
some trace. At the same time there is a growth 
inwards, so that in some cases the bone is reached. 
On the ulna the marks of the great wing feathers are 
easily discernible. 
The cap found on the top of young feathers is 
formed from the outermost cells of the epidermis, the 
