PROFESSOR G. G. STOKES ON THE COEFFICIENT OF VISCOSITY OF AIR, 789 
In Maxwell’s experiments a was 5'28 inches, and when the fixed and movable 
disks were closest h was 0'18475. If we suppose the whole loss of energy 8 per cent, 
greater than that due to rotation round the axis of figure, to which it was deemed to 
be due, we have n— 0‘08, giving 8=1° 8'. Now, no special adjustment was made to 
secure the strict horizontality of the movable disks, or at least none is mentioned ; the 
final adjustment is stated to have been that of the fixed disks, which were presumably 
adjusted to be parallel to the movable ones, and at the desired distance. Hence such 
small errors of level as that just mentioned may very well have occurred. 
Second Note .—On the Effect of the Rotations of the Cylinders or Spheres round their 
own Axes in increasing the Logarithmic Decrement of the Arc of Vibration .— 
By the same. 
(Received October 22, 1886.) 
In Art. 9 of my paper on Pendulums I pointed out that in the case of a ball 
pendulum the resistance due to the rotation of the sphere round its axis need not be 
regarded, on account of the large ratio which the distance of the centre from the axis 
of suspension bears to the radius of the sphere. In Mr. Tomlinson’s experiments the 
corresponding ratio is not near so great, and its squared reciprocal is not small enough 
to allow us to neglect the correction altogether, especially in the case of the spheres, 
the radius of which is much larger than that of the cylinders. In both cases the 
problem admits of solution. 
In both cases the motion of the suspended body may be regarded as compounded of 
a motion of translation, in which the centre oscillates in an arc of a circle, and a 
motion of rotation about its axis of figure, supposed fixed ; and, the motion being 
small, the effects of the two may be considered separately. It is the latter with 
which we have at present to deal. As regards the motion of translation, the spheres 
or cylinders were sufficiently far apart to allow us to regard each as out of the 
influence of the other, and accordingly as oscillating in an infinite mass of fluid ; and 
this is still more nearly true as regards the motion of rotation. The problem, then, 
is reduced to this : a sphere or cylinder performs small oscillations of rotation about 
its axis of figure, which is vertical and regarded as fixed, in an infinite mass of viscous 
fluid ; it is required to determine the motion, and thereby to find the effect of the 
fluid in damping the motion of the system of which the suspended body forms a part. 
In the case of the sphere the problem of determining the motion of the fluid is 
identical with that solved by Professor Von Helmholtz in a paper published in the 
40th volume of the ‘ Sitzungsberichte ’ of the Vienna Academy, p. G07, and reprinted 
in the first volume of his collected works, p. 172, with the exception that the arbi¬ 
trary constants which occur in the integral of the fundamental, ordinary differential 
equation are differently determined, since the condition that the motion shall not 
