Z7'^ TEGUMENTARY ORGANS 



the process, which is called the raphe. These ridges, therefore, bound as 

 many grooves which branch off from the medio-dorsal groove, becoming 

 gradually shallower to the raphe. These secondary grooves, as they 

 might be termed however, are not themselves simple ; their walls, the 

 ridges, being again produced into short parallel laminae, and therefore 

 giving rise to tertiary grooves, branching off from the secondary ones. 

 Now, the whole surface of the matrix being covered by an ecderonic 

 layer in process of conversion into the cortical and medullary substances 

 of the feather, the primary groove becomes filled by the end of the 

 shaft ; the secondary grooves by the terminal barbs, the tertiary 

 grooves by their barbules, while the processes appear to be out- 

 growths from these. Were all this conical horny cap to remain 

 entire, the result would be a very complex sort of porcupine's quill ; 

 instead of this, however, it breaks up along the line of each ridge, and 

 so we have a feather. 



The extremity of the feather being thus constituted, how is its 

 remaining length developed ? According to Reichert, the whole pulp 

 elongates, and as fast as a portion of the feather is completed, the 

 corresponding segment of the pulp dries up, constituting for the vane 

 what has been called the inner striated membrane {e'). However, I 

 believe that this is not the case, the inner striated membrane being, 

 like the outer, a mass of cornified cells detached from the surface of 

 the pulp, just as we shall see the pith of the shaft to be, though this has 

 been also declared by Reichert to be dried-up pulp. I believe that the 

 growth of the feather, on the other hand, resembles that of the hairs 

 and nails ; viz. the extremity as it is finished, is pushed up by the 

 growth of the base, the pulp only supplying materials from its surface ; 

 and I account for the inner striated membrane by supposing that a 

 comparatively imperfect development of horny cell membranes takes 

 place from that surface of the pulp which would otherwise be left bare, 

 when the terminal cone or plume of the feather is pushed away. 

 When the development of the shaft has gone on in this manner for a 

 longer or shorter time, according to the length of the feather, a change 

 takes place. The primary groove, which has gradually widened with 

 the width of the shaft (to the exclusion of the secondar}- grooves, 

 which gradually shorten and ultimately disappear) becoming 

 shallower, extends all round the pulp, and the formation of medullary 

 feather substance ceases, that of cortical substance alone remaining. 

 Thus is the hollow quill formed, and its edges, not quite closing above, 

 leave the minute umbilical aperture by which the inner striated 

 membrane is continued into the " pith " of the quill. This pith 

 is produced b}' the throwing off of successive transverse horny 



