24 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 48 
processes of the latter. The incorrectness of this statement can 
easily be demonstrated by the examination of a surface view by 
which it is to be seen that in their natural position the hooks lie 
some distance beyond the zone of these ventral processes and in 
reality grasp the more proximal, band-like portions of the curved 
fibers. They are moreover too short to reach as far down as the 
ventral margins of the curved fibers from which these processes 
protrude horizontally. Finally, one must consider that if the hooks 
actually gripped these processes they would not be capable of gliding 
backward and forward, but would be wedged in firmly between them. 
According to my own observations the hook-fibers overlie the curved 
fibers and cross them at an angle of nearly go° (figs. 17, 18, 19). 
The hooks hanging down cling by their lower margin to the in- 
voluted upper margin of the curved fibers. All the hooks of one 
hook-fiber hold different curved fibers (figs. 24, 25), so that as many 
curved fibers are held by one hook-fiber and as many hooks grip 
each curved fiber as there are hooks on each hook-fiber. The hooks 
can glide backward and forward in the smooth groove of the upper 
margin of the curved fiber, a fact which is of great importance in 
flight. During each downward wing stroke a pressure is exercised 
on the lower side of the feather-plate. As Parseval (1880, p. 70) 
asserts, the angles between the secondary and the primary quills 
grow more obtuse under the action of this pressure, and the distance 
between the outer part of the secondary quills increases. At the 
same time the feather changes its curvature and its surface in- 
creases. These changes are made possible by the gliding of the 
hooks which connect the two systems of tertiary fibers. If the sec- 
ondary quills move apart, the hooks glide from the middle of the 
ribbon-like portion of the curved fibers which they clasp during 
rest, towards their distal end (fig. 19. See the direction of the 
arrow). If the pressure were still increased the hooks would glide 
still further and slip off the curved fibers, thus losing their hold 
altogether if it were not for the peculiar arrangement for arresting 
them, which effectually prevents this. The bending of the curved 
fiber itself resists such a slipping off and the dentate protuberances 
of this part of their upper margin (fig. 28, sf) described above, 
make it still more difficult. These teeth being directed backward 
it becomes easy at the same time for the hooks to glide back into 
their proper position, should they have been carried beyond by some 
unusual force. 
