40 



Mr. H. Tomlinsoii. 



[Jan. 14, 



IV. "The Coefficient of Viscosity of Air." By HERBERT 

 TOMLINSON, B.A. Communicated by Professor G. G. 

 Stokes, P.R.S. Received January 6, 1886. 



(Abstract.) 



The author has had occasion, whilst investigating the internal 

 friction of metals, to determine the coefficient of viscosity of air. 

 The viscosity of air has already engaged the attention of several dis- 

 tinguished experimenters, amongst others, of G. Gr. Stokes, Meyer, and 

 Clerk Maxwell. The results obtained, however, differ so widely that 

 it was considered necessary to institute fresh researches into the same 

 subject. 



The author employed the torsional vibrations of cylinders and 

 spheres, suspended vertically from a horizontal cylindrical bar, and 

 oscillating in a sufficiently unconfined space. The bar was suspended 

 by a rather fine wire of copper or silver attached to its centre, which, 

 after having been previously subjected to a certain preliminary treat- 

 ment with a view of reducing the internal molecular friction, was set 

 in vibration. The vibrations were performed in a large box, which 

 was rendered sufficiently air-tight to prevent currents of air from 

 vitiating the results. The wire, which was about 97 cm. in length, 

 was suspended in an air-chamber, the double walls of which enclosed 

 between them a layer of water. This air-chamber was in turn 

 surrounded by a second, also provided with double walls which con- 

 tained sawdust in the space between them. The object of the two 

 air-chambers was to protect the wire as much as possible from small 

 fluctuations of temperature, which last had been found to render the 

 internal friction of the metal very uncertain. 



The coefficient of viscosity of air was obtained from observations of 

 the diminution of the amplitude of vibration, produced by the resis- 

 tance of the air to the oscillating spheres or cylinders attached to the 

 horizontal bar, arrangements having been made so that the vibration- 

 period of the wire should remain the same, whether the cylinders or 

 spheres were hanging to the bar or not. In deducing the value of 

 the coefficient of viscosity from the logarithmic decrement, the 

 author has availed himself of the mathematical investigations of 

 Professor G. GL Stokes* 



Five sets of experiments were made with hollow cylinders and 

 wooden spheres, in the construction and measurement of which con- 

 siderable care was taken. When the cylinders were used arrange- 



* See Professor Stokes's paper " On the Effect of the Internal Friction of Fluid* 

 on the Motion of Pendulums," "Trans. Camb. Phil. Soc," vol. ix, Part II, 1850. 



