Current of Electricity from a hot Platinum Wire, etc. 509 



The above results can be fully explained if it is assumed 

 (a) That the negative discharge is chiefly due to the ion- 



isation of gases occluded by the metal, and is for any 



given temperature proportional to the amount of gas 



in the wire. 

 (6) That when the wire is heated, the occluded gases 



slowly escape, the rate of escape diminishing with 



continued heating. 



(c) That the disintegration of the metal is due to the 



carrying away of particles of the metal by the gas 

 as it escapes. 



(d) That a small fraction of the discharge is carried by the 



molecules of gas and particles of platinum emitted 

 by the incandescent wire. 



That some relation exists between the amount of gas in the 

 wire and the disintegration follows from the fact that, although 

 the discharge at one temperature may be sixty times as large as 

 that at a lower temperature, yet the proportion of corpuscles and 

 heavy particles is almost the same in the two cases. This is well 

 shown by the results for temperatures 940° C. and 1090° 0. in 

 Table VII., and affords evidence for assumption (c). 



In general, however, it is noticed that as the heating is 

 prolonged the discharge contains fewer and fewer heavy particles. 

 This is quite in accordance with the above views. For in a mass 

 of platinum, there will be a number of particles much more 

 detachable than the rest. Thus after continued heating the 

 escape of a certain quantity of gas from the wire will produce less 

 disintegration than would be produced by the escape of the same 

 quantity of gas when the wire was heated for the first time. 



In conclusion, I take this opportunity to express my best 

 thanks to Professor Thomson for many valuable suggestions 

 during the course of the experiments which were carried out 

 at the Cavendish Laboratory. 



