34 Natural and Artificial Flight. (January, 
M. Marey’s decision to still further investigate, and to depend 
on experimental results alone. The laws that might be de- 
duced involve in their deduction such complicated calcula- 
tions that it would be easy for an errorto creep in. Besides, 
possessing the means by which distant motions can be regis- 
tered, or rather caused to register themselves automatically 
throughout their whole amplitude, the experiment becomes 
a matter of no great difficulty. The problem to be solved 
is that of so arranging the membrane of the apparatus 
shown in Fig. 1 as to record the bird’s rise and fall in a 
vertical plane,—that is to say, the rise and fall of the bird 
should produce a varying pressure on the membrane of the 
metallic drum. Supposing that a flying bird carries on its 
back a drum of the form described, with the membrane up- 
permost,—if this membrane follows the motion of the bird 
there will be no displacement of the air in the apparatus, 
and the style will remain motionless. But if the membrane 
is prevented from following the motions of the bird, if there 
can be given to it a tendency to remain at rest, while the 
body of the drum attached to the bird is moved, the air in 
the apparatus will be compressed or rarefied, and motion 
imparted to the registering style. This tendency can 
clearly be produced by loading the membrane with an inert 
body, such as a disc of lead. Fig. 5 illustrates this; the 
drum is seen attached to the registering apparatus, with the 
inert mass, the disc of lead, upon the membrane. Practi- 
cally there are several discs, which may be added or removed 
as the exigencies of the experiment require. The move- 
ments in the horizontal direction are not registered, because 
elevation and depression can alone affect the inertia of the 
leaden discs, for when the bird ascends the inertia of the 
mass resists the upward movement, and causes a record 
similar to that which would have taken place had the mem- 
- brane itself been subjected to pressure, and the drum had 
remained motionless. To prote¢t the apparatus from acci- 
dental pressure by the wings of the bird, a wire cage was 
employed. There is, however, still the objection that an 
inert mass placed on an elastic membrane may execute 
vibrations peculiar to itself, and that the apparatus will 
record these vibrations, as well as the oscillations of the 
bird. How can this complication be removed? ‘The law 
of vibrations teaches that the greater the mass, and the 
feebler the elasticity, the longer will be the period of vibra- 
tion. Then as the motions we are studying are tolerably 
frequent, if we arrange matters so that the vibrations of the 
disc shall be of a longer period than the oscillations of the 
bird, we get rid of the difficulty. 
