MOVEMENTS OF THE WINGS OF BIRDS. 249 



turns, would give lateral movements to the pivot on which it 

 rests, and to the base itself. 



To furnish the bird with a solid point of suspension, pro- 

 tected not only from vertical oscillations, but from move- 

 ments of torsion, we have placed at each end of the instrument 

 a cross piece of wood, to the two extremities of which are 

 attached cords communicating with the ceiling of the room. 

 At this point is a revolving hook, which turns freely with the 

 machine. 



Of the apparatus which suspends the bird. — Fig. 104 shows 

 the details of this suspension which binds the bird to the arm 

 of the instrument, while it confines as little as possible the 

 liberty of its movements. 



Of the registering apparatus. — The transmitting tubes are 

 arranged along the arm of the instrument ; they are fastened 

 to it throughout all its length, and end in a register which 

 carries three lever-drums tracing on the revolving cylinder. 

 The instrument in its rotation would cause the transmitting 

 tubes to roll round its axis, if the register to which they are 

 directed did not participate in the general rotation. 



We see in fig. 103 how this apparatus is arranged. The 

 cylinder is placed vertically above the axis of the instrument ; 

 the three levers trace upon it. The whole apparatus rests 

 on a tablet, which turns on the central pivot. We have here 

 well-known arrangements, in which several movements are 

 registered at the same time on the cylinder ; it will, there- 

 fore, be useless to repeat the precautions which are to be 

 taken in the management of the apparatus, such as the exact 

 superposition of the tracing points, &c. 



The movements of the wing are extremely rapid ; they can 

 be registered only on a cylinder turning with great velocity ; 

 that which is employed in this experiment makes one revolu- 

 tion in a second and a half. The shortness of the time at 

 our disposal to trace the movements of the bird compel us to 

 do so only at the precise moment when the phenomena which 

 we wish to observe are presented, whether it be the swiftest 

 flight, the gradual slackening of its speed, or the efforts 

 made at starting. If the three levers were to rub constantly 

 on the cylinder, we should soon have nothing but a confused 

 12 



