38 



Experiments on Fluid Viscosity. [Nov. 21, 



VII. "Experiments on Fluid Viscosity." By A. Mallock. 

 Communicated by Lord Kelvin, P.R.S. Received July 

 26, 1895. 



(Abstract-) 



The paper gives an account of a series of experiments on the vis- 

 cosity of water made with, revolving cylinders. The cylinders were 

 coaxial, and the fluid was placed in the annulus between them. 

 Special precautions were taken to eliminate the effect of the terminal 

 conditions at the lower end of the annulus, with the result that the 

 motion of the fluid in the annulus was practically the same as it 

 would have been in an annulus of infinite length. 



One of the cylinders was driven at known and constant speeds, 

 and the moment transmitted to the other cylinder, which was sus- 

 pended by a torsion wire, was measured for each speed. In some of 

 the experiments the inner cylinder was suspended on the torsion 

 wire, and the outer one revolved, and in others these conditions 

 were reversed. Two sets of cylinders were used, giving annuli 

 of one inch and half an inch width respectively. When the outer 

 cylinder was the revolving one, it was found that, up to a certain 

 speed, the transmitted moment increased as the first power of the 

 speed. Above this speed there was a region where that moment was 

 indeterminate between certain limits, and varied suddenly from one 

 value to another at uncertain intervals. At higher speeds again the 

 moment was perfectly determinate, but varied as a power of the 

 speed higher than the square. 



When the inner cylinder was the revolving one, none of these vari- 

 ations in the character of the function representing moment in terms 

 of speed were observed. At not even the lowest speeds was the 

 moment proportional to the first power of the velocity, but the whole 

 series agreed fairly with the equation, moment — constant x velo- 

 city to the power of 1*8. 



Returning to the experiments where the outer cylinder revolved, 

 the coefficient of viscosity of water deduced from these experiments 

 at speeds where the moment was proportional to the first power of 

 the speed, was greater than Poiseuille's value, and the difference was 

 greater with the one inch than with the half inch annulus. In some 

 former experiments of the same kind where an annulus of about 

 0*2 in. was used, the value of the coefficient of viscosity obtained 

 was very nearly the same as Poiseuille's. 



Taken together, these experiments tend to show that there must be 

 more than two forms of flow possible for a viscous fluid subjected to 

 shearing force, and that long before anything like permanent eddying 



