so that it may be bent downward very easily by pressure near the tip, but does not bend so easily the other way. Notice, too, that the shaft is not straight, but bent so that the upper surface of the feather is convex, and the lower concave. We shall soon find a meaning in all these properties.
Passing now from the shaft to the vane, we see that the two sides are not alike: one edge, or vane—the one on the right-hand side of every feather in the right wing, that is, the side nearest the tip of the spread wing—is short and stiff, while the vane on the other side of the shaft is broad and flexible; and, by examining the feathers in their places in the wing, we see that the broad inside edge of the first or outside feather passes under the stiff, narrow outside edge of the next feather, which has its own inner edge supported in the same way by the third feather, and so on through the wing. When the wing is flapped downward through the air, the broad edge of each feather is pressed against the inflexible narrow edge of the feather next it, and the whole wing is thus made air-tight; but, when the wing is moved the other way, the broad edges have nothing to support them, and are pressed downward, so that the air can pass between the feathers.
The vane is also made up of separate pieces. If it is carefully examined, separate pieces, or "barbs," will be seen running off from each side of the shaft at a slight angle, and parallel to each other, united in such a way as to form two flat plates, the vanes. These barbs are fastened to each other quite firmly, but, if part of the vane is pulled down toward the quill, the barbs will separate at last with a tearing sound, and if this is repeated in a few places it will give the feather a very draggled appearance, and it will seem torn beyond possibility of restoration; but, if it be drawn gently between the fingers two or three times from base to tip, the broken places will unite so perfectly that it may be quite impossible to find them again. The working of the mechanism by which the attachment is made is so perfect that it need only be noticed to be admired, and careful examination will reveal the simple means by which it is accomplished.
Each barb, when examined with a lens, is seen to bear some resemblance to the whole feather; like the feather, it has a shaft running longitudinally, and a vane on each side of it. These vanes are unequal, as in the whole feather, and they are composed of separate pieces running off from the shaft, and called "barblets," because they are to the barb what the barbs are. to the whole feather. On the side of the barb toward the tip of the feather the barblets run out from the shaft of the barb nearly at right angles, and send off from their lower surfaces little hooks at regular intervals, all pointing downward; on the other side of the barb the barblets have no hooks, and, instead of being set at a large angle with the shaft, they are almost parallel with it, so that where they meet and run under the hooked