CAUSE OF PRODUCTION OF "DOWN." 169 



fibrous. Fig. 33, PI. III., represents the final stage in develop- 

 ment. That the formation of this horny tube is wholly different 

 from the process by which the shaft and quill of the definitive 

 feather are formed, as described by Davies (p. 594 et seq.), is 

 evident. Instead of being a process designed for the accomplish- 

 ment of a definite work — the building of shaft and quill — it 

 appears to be due to a lack of differentiation of the cell mass and 

 a short cut to cornification of the tissues induced by a reduced 

 blood supply to this part of the feather during the period when 

 the cells would be showing differentiation if supplied with suffi- 

 cient nourishment. It is significant that this condition of a corni- 

 fied ring instead of the normal barb-vanes is more often found 

 among the strictly altricial birds which are hatched in a helpless 

 condition. It is well known that the first few days after the 

 hatching of altricial birds are the most critical days of their lives. 

 During this critical period there appears to be no growth of the 

 down. An American robin which hatched on the fourteenth 

 day of incubation possessed the usual down upon the head and 

 back. These downs made no further growth. It was not until 

 the fourth day after hatching that the skin gave evidence of the 

 beginning of the definitive feathers. On the eighth day after 

 hatching the skin surface was exposed to the drying influences 

 of the air before renewed activity in the feather germ began. 

 During this interval of four days the so-called ' quill ' was formed 

 at the proximal end of the down by the rapid drying of the im- 

 perfectly formed barb-vane ridges " (p. 13). 



I wish to make a number of statements concerning the para- 

 graph just quoted ; for, from my point of view, a number of things 

 are here touched — but hardly grasped — and, at any rate, not 

 made really clear. 



In the first place the statement is made (here and elsewhere) 

 that in the formation of the " quill " a lack of differentiation is 

 the process at fault. My own studies on similar structures — 

 and I think Jones' plates, as well as some points of his descrip- 

 tion show -the same — indicate that while lack of differentiation 

 is undoubtedly a part of the process, a more important part is 

 lack of growth. The barbules, for example, do not differentiate, 

 but the cells which should form them do not grow. That is to 



