MASCHA] STRUCTURE OF WING—FEATHERS 13 
is continued in a distal part which is more rod or thread-like, and 
from which numerous projections arise ; the proximally ventral hooks 
characteristic of these fibers, and the distally ventral and dorsal 
spines lying in pairs opposite each other. 
We will first consider the hook-fibers of the remiges of Columba 
livia as a typical example (figs. 2, 26). Here the proximal band- 
like portion forms about half of the whole fiber. It is transversely 
bent, its upper half being vertical, the lower turned obliquely toward 
the secondary quill and the front. As transverse sections clearly 
show, the fiber is here not only groove-shaped, but thickest at its 
upper margin, decreasing in thickness downward and passing finally 
into a thin ventral membrane. Only at the proximal end of the 
fiber the ventral membranous extension is absent. In studying trans- 
verse sections through the secondary quills (figs. 3, 4, 26) we see 
that in the first two the hook-fibers are thickest at the base and 
become thinner towards the top; in the third section the fiber is 
nearly equally thick at the top and the bottom and in the next sec- 
tion the reversed relation, holding good for the remainder of the 
hand-like part, sets in. At the upper margin a distinctly recogniz- 
able swelling (figs. 8, 26, WIt) is to be seen, which generally dis- 
appears distally, where the lower part of the fiber band passes into 
the thin ventral membrane mentioned above. The dorsal swelling 
is sometimes, particularly when pigment is present, plainly visible 
even in surface views. The histological structure of this part of 
the hook-fiber is as follows: If we look at a hook-fiber from the side 
we sometimes notice in its proximal half a row of oval spots usually 
with dark margins, their long axes extending obliquely upward and 
backward. These spots were first noted as depressions, later authors 
recognized them as dried-up nuclei (fig. 9, k). Schroeder (1880, 
p. 30) was the first to state that each fiber consists of a single row 
of cells. This supposition found confirmation in the subsequent 
investigations of Klee (1886) and Davies (1889). 
The nuclei of the cells forming the fiber lie in its thinner portion 
and are clearly visible. Apart from the nuclei, one can frequently 
perceive the cell limits also, either as fine, dark lines, as I found them 
especially in the hook-fibers of Cypselus apus (fig. 16, Zgr) after 
staining, or as clear, pigmentless lines separating the upper por- 
tions of the cells in which the greatest amount of pigment is con- 
tained (figs. 8, 9, 10). Strong (1902, p. 156) has fully described 
these cell limits and has drawn particular attention to the fact that 
they first extend from the upper margin obliquely forward and 
downward and suddenly turn in the proximity of the nuclei, being 
