288 Mr Richardson, On the Negative Radiation 



than 1*5 amperes ; it was found that the temperature of the wire 

 was practically constant during each observation ; the variation 

 (about 5°) being due to gas given off from the wire. The tem- 

 perature of the wire therefore steadily decreased during each 

 observation, but it was held that by taking the mean of readings 

 immediately before and after each measurement of current from 

 the wire the true temperature would be obtained. The resistance 

 was measured by placing the wire in one arm of a Wheatstone's 

 bridge, the corresponding arm being a thick German silver 

 resistance of 1*7 ohms. This resistance was not sensibly heated 

 by the current which passed through it. The resistance of the 

 other two arms was of the order of 1000 ohms, so that practically 

 the whole of the heating current passed through the wire and the 

 German silver resistance. The heated platinum wire passed 

 axially through the aluminium cylinder, being fixed to two 

 electrodes in the bulb which contained the cylinder. The bulb 

 was connected with a drying apparatus, pump and McLeod gauge. 

 In most of the experiments the pressure was about *02 mm. but 

 it varied from "01 to "16 mm. It was very difficult to keep the 

 pressure down at the higher temperatures owing to the gas given 

 off by the hot wire. 



The same Thomson galvanometer was used both to measure 

 the leak and for the Wheatstone's bridge. It had a resistance of 

 4058 ohms and gave 1 scale division for a current of 7 x 10 -10 

 amperes. In order to use it for both purposes, suitable shunts 

 had to be inserted. The whole of the measuring apparatus was 

 insulated in paraffin so that the potential of the hot wire could be 

 raised to any multiple of 40 volts up to 400. It was found that 

 with 400 volts positive on the wire, there was no observable 

 deflexion of the galvanometer, whereas quite big currents were 

 obtained when the wire was negative. In all cases the saturation 

 current, i.e., the total number of ions given off by the wire, was 

 measured. 



To reduce the determinations of resistance of the platinum 

 wire to temperatures use was made of the measurements of the 

 melting points of potassium and sodium sulphates by Messrs 

 Hey cock and Neville 1 . The wire was set up in air and its re- 

 sistance determined first at the ordinary temperature, and after- 

 wards when the smallest possible grain of potassium sulphate 

 placed on it just melted. In this way the resistance for two 

 temperatures differing by about 1000 degrees was obtained, and 

 the temperature corresponding to any other resistance reading 

 could be got by interpolation from the curves given by Professor 

 Callendar 2 . To test the method, the melting point of sodium 



1 Heycock and Neville, Chem. Sov. Journal, lxvii. 1895, p. 160. 



2 H. L. Callendar, Phil. Mag. xlviii. 1899, p. 519. 



