48 Dr. A. 0. Rankin e on the Relation between 



Table II. shows the numbers relating to the curve. 

 Table II. 



Gas. 



T c . 



A. 



»/ c Xl0 4 . 



V C XW 



(calculated). 



Difference 

 per cent. 



He 



(5)? 



3-96 



0021 



0-32 





Ne 





20-03 









A 



155-6 



39-92 



1-261 



1-253 



06 



Xr 1 



210-5 



83-0 



1-828 



1-806 



12 



X 



288 



1307 



2-220 



2266 



1 



-21 



The numbers in the fourth column are calculated from 

 Sutherland's equation using- rj , C, and T c ; those in the fifth 

 column are deduced from the equation 



2 



q-=3-93xl0- 10 . 

 A 



With the exception of helium this equation is a good fit, 

 and even the exception seems capable of reasonable ex- 

 planation. In the cases of helium and argon we may avoid 

 the necessitv of such extensive extrapolation by using the 

 experimental results recorded by Schmitt *. These measure- 

 ments extend from — 193°*2 C. to -fl83°'7 C, and the value 

 of the viscosity of argon at the critical temperature can 

 be deduced by interpolation. The value thus obtained is 

 1*25 x 10~ 4 , which is remarkably near the value calculated 

 here, and suggests that in this case the extrapolation is valid. 

 When we turn to helium, however, the results are very 

 different. At a temperature of 80° absolute the actual value 

 of the viscosity exceeds that calculated by using Sutherland's 

 equation by 26 per cent., and the divergence increases with 

 fall of temperature. The temperature at which the viscosity 

 is required is 5° absolute, or 75° lower, where we should 

 expect the divergence to be more serious still. In fact, an 

 extrapolation over this range of 75 degrees indicates that 

 032 x 10~ 4 is a much more probable value of the actual 

 viscosity at 5° C. than 0*021 x 10~ 4 . This would bring 

 helium into line with the other gases. It is quite possible, 

 therefore, that helium also conforms with the law here pre- 

 sented, and that the apparent divergence should be attributed 

 to the failure of Sutherland's equation at temperatures so 

 near to absolute zero. 



* Ann. tier Physik, Bd. xxx. Heft 2, p. 399 (1909). 



i 



