FLYING-MACHINES, ETC. 457 



tween hovering and the ordinary flight of progression, while the am- 

 plitude of the changes in the plane of the extremity of the wing is 

 essentially a function of the velocity of translation of the bird. At 

 the extremity of the wing, where the most considerable changes of 

 plane takes place, these changes equal 90, and even more, during hover- 

 ing ; but then displacements of plane are far less in the flight of pro- 

 gression. According to our calculations the extreme portions of the 

 surface of the terminal feathers of the ci'ow's w'ing are, during free 

 flight, inclined forward during the depression of the wing only from 

 7 to 11 below the horizontal, and from 15 to 20 above the hori- 

 zontal plane during the elevation of the wing. The plane of the 

 wino; at its base acts durino- the above motions like a kite inclined at 

 an angle only of from 2 to 4. 



It is easy to verify the slight inclination of tlie wing, and conse- 

 quently the smallness of its angles of action in the air, by observing a 

 flying bird moving in an horizontal line of sight, for we then see only 

 the edges of the wings. It is, in short, inexact to say that the wing 

 changes its jilatie ; we can barely say that it changes its j^lcincs. The 

 trutli is, that it is gradually more and more warped in going from its 

 base to its extremity. It was so understood, indeed, by an English 

 author, whose labors we became acquainted with after we had con- 

 structed our bird, and to him we are indebted for liaving saved us 

 several researches. The theory of Sir G. Cay ley, published in 1810, 

 difiers from ours but in a few particulars. He is of the opinion that 

 the outer portion of the wing in ascending exerts always a propulsive 

 action, and lie atti'ibutes to the propelling parts and to the sustain- 

 ing, kite-like parts of the wing, proportions which are relatively the 

 reverse of those to which w^e have been led by our calculation. 



It was with these ideas, favorably judged of by the Academy in 

 September, 1871, that we undertook the application of the torsion of 

 caoutchouc to the problem of the mechanical bird. The wings of our 

 bird are made to beat in the same plane by means of a crank and con- 

 necting- rods. After several rouefh trials, we found out that the trans- 

 formation of motion in the machine required a mechanism very solid 

 relatively to its weight, and I requested M. Tobert, an able mechanist, 

 to construct out of steel a piece of mechanism designed by my brother, 

 E. Penaud. The accompanying figure represents the apparatus so 

 constructed ; C C is the motor of twisted caoutchouc placed above 

 the rigid rod, PA A, which is the vertebral column of the machine; 

 from this rod, at A and A, ascend two rigid forks, which serve below 

 as supports for the crank, C H, which is attached to the twisted 

 caoutchouc ; and above, at the ends of the forks at and 0, are the 

 pivots on which the wings oscillate. The links, H S, convert the 

 motion of rotation of the crank into the reciprocating motion of the 

 arms, MZi, O M L. At ^ is a steering-tail, which we found by 

 experience was best made from one of the long feathers of a peacock's 



