34 The Bird 



upward and escape beyond the edge. So it would, if it 

 were not for the arrangement of the feathers on the wing, 

 which overkij) Hke the tiles on a roof, each vane over- 

 h'ing and holding down the long barbs of the feather 

 in front, while, above and below, other shorter feathers 

 help to bind the whole tight h', thus enabling the bird at 

 ever}^ stroke to whip a wingful of air downward and 

 backward. 



A feather and its parts, like all the rest of the bird, is 

 composed of cells — empty and hollow ones in this in- 

 stance, as we can easily see for ourselves by placing a 

 barb from a pigeon's feather in a drop of water and 

 looking at it under a low-power magnifying-lens. The 

 network of horny cells is yery plain. 



It is a simple matter to say that a feather consists of 

 quill, barb, barbules, etc., but to appreciate the wonder- 

 ful complexity of this structure let us make a little cal- 

 culation. Suppose we have a wing-feather from a com- 

 mon pigeon with a vane about six inches long. If we 

 have patience enough to count the barbs on one side of 

 the quill, we will find there are about six hundred of 

 them. So the vane of the entire feather has twelve 

 hundred of these little side featherlets. One of these, 

 from a narrow part of the vane, will shovv under the micro- 

 scope about two hundred and seventy-five pairs of bar- 

 bules, which multiplied by the number of barbs on that 

 side amounts to three hundred and thirty thousand. 

 Making a verj^ low estimate of the whole vane, we have 

 nine hundred and ninety thousand separate barbules on 

 this one feather, and when we think of the innumerable 



