640 Miss J. M. W. Slater on the 



II. Influence of Heat on the Excited Activity. 



In view of the above results, it was thought interesting to 

 investigate the effect of heat on the excited activity in more 

 detail than had previously been attempted, in order to obtain 

 if possible a further separation of the two stages by a kind 

 of fractional distillation. Von Lerch*" has investigated the 

 amount of activity removed by heating for a short time at 

 different temperatures, fie found no loss of activity at 800°, 

 but 16 per cent, was removed by heating for half a minute 

 at 1020°, and the rate of volatilization increased as the tem- 

 perature was raised to 1460°, at which point 99 per cent. 

 was removed after 1-J minutes total heating. He does not 

 seem to have observed the rate of decay of the activity 

 remaining on the wire after heating at the lower temperatures, 

 so that his results give no indication as to the possibility of a 

 separation by this means. 



Method of Investigation. 



The method used was as follows : — A platinum wire was 

 made active by negatively electrifying over thorium in the 

 usual way ; its activity was tested by the same apparatus as 

 that used for the active disks in the previous investigation. 

 The wire was heated by an alternating current obtained from 

 a transformer coil. The temperature could be regulated by 

 varying the number of turns of wire on the transformer, 

 while for fine adjustments a rheostat of stout german-silver 

 wire dipping in mercury was used. The platinum wire was 

 placed along the axis of a glass tube, to protect it from air 

 currents, and its temperature was measured by a thermo- 

 couple of fine platinum and platinum-rhodium wires. The 

 galvanometer deflexions were standardized by observing the 

 melting-point of very small crystals of pure sodium sulphate 

 placed on the hot wire close to the junction, and then using 

 Callendar's curve of corrections f. As it was not possible to 

 have the junction welded to the wire, the readings may have 

 varied somewhat, according to the way in which the junction 

 lay on the wire. The junction was held down by a spring, 

 and care was taken that its position should, be as nearly as 

 possible the same in each experiment. In standardizing the 

 galvanometer readings a number of independent observations 

 were taken, in which the position of the junction and the 

 tension of the spring varied in the same way as in the ex- 

 periments with the active wire. The deflexions corresponding 



* Drude's Annalen, Nov. 1903. 



t Phil. Mag. vol. xlviii. 1899, p. 533. 



