358 On Thermal Radiation in Absolute Measure. [June 16, 



the wire heated by an electric current, and determining in absolute 

 measure the electric energy necessary for this purpose. This energy 

 is lost in radiation and in conduction at the ends, but chiefly in 

 radiation. The wire is contained in a long copper tube, blackened 

 inside, and kept cool by a water jacket ; and the surface of the wire 

 may be bright and polished, or may be modified by being coated with 

 lamp-black, platinum-black, oxide of copper, or some other material. 

 Polished wires have been chiefly used hitherto ; but arrangements 

 for comparison of wires with surfaces differently prepared are 

 described in the paper. 



Two methods of determining the electric energy have been used. 

 One consists in measuring the electric current and the difference of 

 potentials between chosen points in the radiation- wire ; the other in 

 measuring the current and determining simultaneously the electric 

 resistance, by means of a Wheatstone bridge, modified to suit the 

 necessities of the case. . 



A knowledge of the resistance of the radiation- wire gives also its 

 temperature, by means of separate determinations (described in the 

 paper) of the law of alteration of electric resistance with temperature. 

 The temperatures of the radiation-wire and of the envelope are all 

 referred to the air thermometer. 



In order to vary the air-pressure surrounding the radiation-wire, 

 and thus obtain data for the purpose of eliminating the heat-carrying 

 properties of the air or other gas, the copper tube is connected, in a 

 manner described in the paper, with a five-fall Sprengel pump ; 

 and the pressures in the extreme vacuums are measured by the 

 M'Leod gauge. 



The results of the investigation, so far as it has gone, are shown in 

 a series of tables and curves. 



A long and very complete series of determinations has been made 

 of the radiation at various given constant pressures, but at different 

 gradually increasing temperatures. 



By means of curves of this kind, showing radiation in extreme 

 vacuum, a comparison may be made between the results of experiment 

 and the results calculated from Stefan's well-known fourth power 

 law. The experimental results do not appear to give any support to 

 that law. 



Several series of determinations have been made at different con- 

 stant temperatures, the pressure being continuously diminished. This 

 mode of experimenting, by far the most appropriate to the purpose in 

 hand, has only recently become convenient to the author, as it requires 

 a special suitable current galvanometer. Further experiments are to 

 be carried out with this method. 



In the meantime it may be said that on continuously diminishing, by 

 means of the Sprengel pump, the air surrounding the wire, a point 



