HOGG. — VISCOSITY OF AIR. 

 TABLE V. p - 10.014 cm. 



623 



A reference to the description of the apparatus will show that the dis- 

 tance between the fixed aud moving surfaces, viz., the inner surface of 

 the glass bulb and the surface of the sphere, was about G.o cm., so that no 

 correction need be applied for slip. 



In addition to what has been said with regard to the neglect of the 

 squares of velocities it may be urged that, if it can.be taken as satisfac- 

 torily shown that the value of (i is constant over as large a ranire of 

 pressure as that over which the foregoing experiments have been made, 

 there is, in the results obtained at the lower pressures, another evidence 

 of the justifiable neglect of quantities of the order of V^ap/fx. For a 

 diminution of p diminishes this quantity while not affecting p., so that, if 

 the value of p. derived from a formula obtained by neglecting the squares 

 of small quantities is the same at atmospheric pressure as it is at a much 

 smaller pressure, 2.5 cm., it may be safely concluded that the use of the 

 formula at atmospheric pressure was justifiable. 



The mean of the results obtained at atmospheric pressure is 



/xo = 0.0001713. 



This is very nearly the mean of the results obtained by the more recent 

 investigators, some of whom used the transpiration method aud others 

 the oscillation method. 



