1894.] On determining the Thermal Conductivity of Metals. 199 



amount of carbon and manganese being added subsequently in the 

 forms of grey iron, spiegel, or ferro-manganese. 



I propose to continue this work by extending my observations to 

 the flame from the basic Bessemer process and the gases in the Siemens 

 steel furnace. 



VI. " On a Method for determining the Thermal Conductivity 

 of Metals, with Applications to Copper, Silver, Gold, and 

 Platinum." By JAMES H. GRAY, M.A., B.Sc., 1851 Exhibi- 

 tion Scholar, Glasgow University. Communicated by LORD 

 KELVIN, P.R.S. Eeceived May 24, 1894. 



(Abstract.) 



The object of this investigation was to obtain a method for deter- 

 mining thermal conductivities of metals, which would not require 

 either elaborate preparations or large quantities of the substances to 

 be tested, and by means of which a test could be made in a few 

 hours. 



The method about to be described was suggested by Lord Kelvin 

 thirty years ago, and is the experimental realisation of the theoretical 

 conditions implied in the fundamental formula 



where the symbols have their usual meaning. 



The apparatus was made so as to be suitable to test the metals in 

 the form of wires of circular section. 



The diameters found most convenient were [from 2 to 4 mm., the 

 lengths from 4 to 8 cm. 



One end of a given length of the wire is kept at a constant known 

 temperature. The rise of temperature of the other end of the wire 

 is noted every minute, and, if proper precautions be taken to prevent 

 loss by radiation from the sides, the data are obtained for calculating 

 the thermal conductivity. 



The wire to be tested is soldered at one end into the bottom of a 

 copper box, 16 cm. long, 6 cm. wide, and 7 cm. deep. The bottom of 

 the box is made of copper 3 mm. thick, the sides of thin sheet copper. 



In the box, immediately above the hole into which the wire is 

 soldered, there is a large block of copper, in which a hole has been 

 made sufficiently large to admit a small thermometer. 



The box is filled with water and supported at its middle by being 

 fitted into an asbestos-lined wooden screen, 24x24 cm. The water 

 is heated by a Bunsen burner placed on the other side of the screen 



