Positive Electricity from Hot Bodies. 



637 



at M was re-established and the current again measured. It 

 was now found to be 176 scale-divisions — that is to say, not 

 very much smaller than the initial leak. The potential- 

 difference of 100 volts was maintained for 10 minutes (with 

 the same temperature as before) and the current noted at the 

 end of this period. This time it was found to be only 

 111 scale-divisions. The discharge from the hot wire was 

 thus reduced to a much greater extent during the second than 

 during the first period. Finally the cylinder was again 

 insulated for a further period of 10 minutes, the temperature 

 of the wire not being allowed to vary. When the current 

 was again measured it was found to be 111 scale-divisions. 



These observations show that the diminution in the positive 

 current from the platinum is not an ordinary fatigue effect, 

 but is due to the decrease in the quantity of ionizable matter 

 which the platinum contains. 



Influence of Water on the Positive Leah from Platinum. — ■ 

 One of the most remarkable phenomena in connexion with 

 the emission of positive electricity by hot platinum is the 

 increase in the discharge caused by the presence of water. 

 I observed this effect by accident during experiments on 

 potassium sulphate. This salt when heated on platinum wire 

 apparently discharged large quantities of positive electricity. 

 It was put on the wire by dipping the latter in a concentrated 

 aqueous solution of the salt. At a temperature of 1100° C. 

 and with a potential-difference of 100 volts the leak was 

 initially far beyond the range of measurement of the gal- 

 vanometer. This leak decayed with great rapidity at first 

 and more slowly and somewhat irregularly afterwards. The 



Fig:. 3. 



§200 

 | 100 





way in which the current varies with the time is shown in 

 fig. 3. As it was found that the same effect could be produced 



