768 
MR, H. TOMLINSON ON THE COEFFICIENT OF VISCOSITY OF AIR, 
Table I. 
Author.* 
Coefficient of viscosity of 
air in C.G.S. units. 
Temperature in 
degrees Centigrade. | 
G. G. Stokes, from Baily’s pendulum 
•000104 
O 
experiments. 
Meyer, from Bessel's experiments 
•000275 
Meyer, from Gieault’s experiments . 
•000384 
Meyer. 
•000360 
18 
Meyer (second paper)f. 
•000333 
8-3 
•000323 
21-5 
. 
•000366 
34-4 
Maxwell. 
•000200 
18 
Further, Maxwell finds the coefficient of viscosity of air to be independent of the 
pressure and to vary directly as the absolute temperature.| The above author gives 
the following formula for finding [x, the coefficient of viscosity, at any temperature 
6 ° C. 
/a =•0001878(1 +*003650). 
Maxwell offers an explanation of the difference existing between his own results 
and those of Meyer, but states that “ he has not found any means of explaining the 
difference between his own results and those of Professor Stokes.” Professor Stokes 
has, however, been good enough to inform me that, as at the time of making his deduc¬ 
tions from Baily’s experiments it was not known that the coefficient of viscosity of 
air was independent of the pressure, but, on the contrary, was assumed by him to vary 
directly as the pressure, the resistance offered by the residual air in Baily’s partial 
vacua was underestimated, and, as a consequence, the deduced coefficient of viscosity 
was too small. It is to be hoped that Professor Stokes will at some future period apply 
the necessary corrections, but as this has not yet been done, and as we have still no 
explanation of the discrepancies existing between the other values of fx given in 
Table I., I wished to make some independent observations on the viscosity of air for 
the purpose of ascertaining how far these would agree with those of Maxwell, in 
which I was inclined to place great confidence. 
Maxwell employed the method of torsional vibrations of disks placed each 
between two parallel fixed disks at a small, but easily measurable distance, in which 
case, when the period of vibration is long, the mathematical difficulties of determining 
the motion of the air are greatly diminished. This method appeared to be a very good 
one, but, as I wished to make my determinations under conditions similar to those 
* For references see Maxwell’s Bakerian Lecture, ‘Phil. Trans.,’ vol. 156, 1866, p. 249. 
f Meyer lias more recently made other determinations of the coefficient, for which see the end of the 
paper. 
t This result does not seem to be confirmed bj 7 other experimenters. (See the end of the paper.) 
