588 



Mr. H. Redmayne Nettleton on a New 



2nd. To test the practicability of using such artificially 

 produced temperature gradients as a means of determining 

 thermal conductivity in absolute measure. 



The author has extended Dr. Griffiths's results so as to 

 take into account radiation, which plays an important part in 

 most cases of conduction, and has carried out the researches 

 suggested. 



(2) Theory. 



Consider a bar moving longitudinally with a constant 

 velocity v as shown and passing through two constant tem- 

 perature sources, and let 



K = thermal conductivity, 

 s = specific heat, 

 p = density, 

 A = cross-section, 

 I = intrinsic heat in volume vA of the bar at the 



temperature of the enclosure, 

 ^ = temperature above the enclosure at distance x from 

 any plane in space normal to the bar. 



Then the quantity of heat crossing a section SS ; in space 

 is due to conductivity -KA-p, and due to the impressed 

 velocity vApsO + 1. 



Fiff. I. 





//<* 





co~e-cU 



Applying this to the small prism abed in space of width 8x 



we have : — 



d6 

 Heat entering ad per sec. = -KA-p -f vAps6 a + I. 



in 



Heat leaving be per sec. = — KAj- 6 + vApsO b -f I. 



In the steady state the difference will be radiated, and 

 since 



b = Q a \ $#7-, 

 dx 



