Effects of Temperature and Pressure on Thermal Conductivity. 337 



" The Effects of Temperature and Pressure on the Thermal Con- 

 ductivities of Bodies. Part I. The Effect of Temperature 

 on the Thermal Conductivities of some Electrical Insulators." 

 By CHARLES H. LEES, D.Sc. Communicated by ARTHUR 

 SCHUSTER, F.E.S. Eeceived November 29, Eead December 15, 

 1904. 



[Abstract.] 



The substance whose thermal conductivity is to be determined has 

 the form of a cylinder about 8 cms. long, 2 cms. diameter, and is 

 surrounded by a thin cylinder of brass, which on account of its com- 

 paratively high thermal conductivity, makes the outer surfaces of the 

 substance isothermal surfaces. The brass cylinder is placed in a Dewar 

 tube. 



The heat is supplied by the passage of an electrical current through 

 a platinoid wire embedded in the substance parallel to the axis of the 

 cylinder, and about '4 cm. distant from it. The amount of heat 

 generated is determined by the current through the wire and the 

 potential difference between its ends. The temperature is measured 

 by the electrical resistance of two short spirals of No. 40 pure platinum 

 wire, down the centre of one of which the heating wire passes. To 

 eliminate errors due to want of symmetry, a second heating wire 

 passes down the centre of the second spiral, and the heating current 

 may be sent through either or both at will. 



The difference of temperature of the two spirals is determined by 

 making them two arms of a resistance bridge, the other two arms of 

 which are equal. By means of mercury cups resistances may be 

 placed in series with either of the spirals till a balance is obtained. 

 From the resistance necessary to effect the balance and the "funda- 

 mental constants " of the platinum thermometers which the two spirals 

 constitute, the difference of their temperatures is calculated. The 

 thermal conductivity k of the medium would, if the latter were infinite 

 in extent and the heating wire infinitely long, be connected with the 

 difference of the temperatures /'i, v- 2 at the two points distant i\ and r* 

 from the heating wire, and the amount of heat H generated per second in 

 1 cm. of the latter, by the equation 



In the apparatus used the connection between the quantities is not 

 so simple, but the theory is worked out completely in the paper, and the 

 corresponding expression given. 



The following fairly representative substances have been tested each 



