Thermal Radiation in Absolute. Measure. 271 



stout multiple platinum terminals ; and by means of these, which are 

 fused, with the help of some white enamel, into the glass at d^d^ the 

 current is passed through the platinum wire. At e,e, platinum wires 

 are brought through the sides of the tube, and serve as potential 

 electrodes; and it is to keep the platinum wire ah in the middle of 

 the length of the tube, and to avoid pulling unduly on the potential 

 electrodes, that the two spiral springs, one at either end of the tube, 

 are introduced. 



Two exactly similar tubes were employed, as shown in fig. 2. They 

 were connected together by a side tube, as shown ; and by means of a 

 branch tube, attached to this side tube and connected to a Sprengel 

 pump, the air was withdrawn from both tubes at the same time ; and by 

 this arrangement it was provided that the vacuum in the two tubes 

 should be at all times precisely the same. 



In one of the tube? the platinum wire was brightly polished and 

 perfectly smooth, just as it came from the maker's hands. The other 

 tube contained a platinum wire cut from the same hank, but with the 

 surface covered with an excessively fine coating of soot. The soot was 

 put on by passing the wire carefully through the upper part of a clear 

 paraffin flame.* 



The usual arrangements were made for drying the vacuum of the 

 tube, and of the pump, by means of phosphorus pentoxide ; and the 

 vacuum was measured by means of the Gimmingham modification of 

 the McLeod gauge. 



The wires were heated, as in the former experiments, by means 

 of an electric current. Fig. 2 shows the electric connections. A 

 battery consisting of a sufficient number of secondary cells was em- 

 ployed ; and ^the current was controlled by means of suitable resist- 

 ances, including a rheostat. In the experiments here described the 

 platinum wires of the two tubes, the resistances, and the battery were 

 all connected in series, so that the same current passed through all.f 



The current in the circuit was measured by means of a Kelvin 

 ampere gauge, and the difi'erence of potentials at the two ends of each 



* The texture of the soot depends greatly on the source from which it is 

 obtained and on the way in which it is applied to the wire. Some preliminary 

 experiments have been made with various coatings of soot, and comparisons have 

 been attemped between surfaces finely coated with soot, and surfaces prepared with 

 platinum black and with a fine coating of black oxide of copper chemically applied 

 to the wire {cf. J. T. Bottomley, ' Phil. Trans.,' A, 1887, p. 449). 



t In another set of experiments the platinum wires were Joined in parallel, and, 

 by means of two rheostats, one connected in series with each platinum wire, an 

 attempt was made to regulate the current in each wire so that the temperatures in 

 the two should be the same. This was found very difficult to carry out ; but it is 

 intended to renew the attempt, and determine simultaneously the radiation from 

 two wires with different surfaces, in the same vacuum, and at the same tem- 

 perature. 



