MR. H. TOMLINSON ON THE COEFFICIENT OF VISCOSITY OF AIR. 
797 
the rod was soldered into the centre of a horizontal, hollow, brass tube, of length 
17'85 centims., of diameter 1'25 centim., and of mass 29'20 grammes. 
From the hollow brass bar the paper cylinder was suspended; two holes, whose 
centres were 2\ centims. from the top, being cut in the walls of the paper cylinder 
for this purpose. 
Great care was taken in arranging the cylinder, so that the axis of rotation might 
coincide in direction as accurately as possible with the axis of the suspended system. 
As the paper cylinder did not quite hang truly, it was made to do so by placing small 
strips of tinfoil, as riders, on the top of the cylinder, and these strips were carefully 
padded down by hand to the walls of the cylinder. The usual previous precautions 
having been taken, the logarithmic decrement was determined from seven sets of 
observations, each involving 100 vibrations, as follows :— 
Number of 
observation. 
Logarithmic decrement. 
1 
•0026307 
2 
•0025856 
3 
•0025837 
4 
•0025810 
5 
•0025700 
6 
•0025849 
7 
•0025550 
These observations were consecutive, and the mean of them is '0025844. 
The paper cylinder was now removed, and in its place was substituted a much 
shorter cylinder, made partly of paper and partly of tinfoil, and having nearly the 
same mass and mean radius. The dimensions of this cylinder were as carefully 
measured as those of the longer cylinder, with both steel tape and calipers. The 
mean of the inside and outside radius was 7'5132 centims., and its real length was 
12'80 centims. Since the radius is, however, not quite the same as that of the 
longer cylinder, we must assume its length to be 
/7'5132\ 3 
12 ' 80X (V569s) CSntimS -’ 
or 12'5 2 centims. if we are to use it for the purpose mentioned below. 
The same pieces of tinfoil as had been used with the long cylinder were used here, 
and for the same purpose. The logarithmic decrement was then determined by six 
sets of experiments, each involving three times the number of vibrations employed 
with the longer cylinder. 
