of the Negative Thermionic Current in a Vacuum. 739 



deflexions of the order of 1 scale-division (1 mm.), too much 

 weight must not be attached to the absolute values of the 

 ratios in the last column of the table. It should perhaps be 

 pointed out that the very small deflexions obtained with the 

 magnetic field do not prove conclusively the presence of 

 heavy ions in the discharge. Similar results would be 

 observed if a few electrons were shot out of the wire with a 

 considerable velocity, or if a few rebounded on to the insu- 

 lated electrode after striking the earthed gauze cylinder. 



It remains to consider why heavy ions to the extent of 

 5 per cent, and more were apparently detected in some cases 

 in the earlier work on this subject (Owen, Proc. Camb. Phil. 

 Soc. loc. cit.). In the experimental tube used then, the insu- 

 lated electrode consisted of two vertical brass disks, 8 mm. 

 apart, and connected together by a single thin bar of brass 

 screwed into them at the extreme end of horizontal diameters. 

 The hot wire was situated midway between the disks, and 

 the magnetic field was arranged so as to deflect the electrons 

 along a vertical diameter. Owing to non-uniformity of the 

 electric and magnetic fields, it is possible that when the 

 thermionic currents were large (as at high temperatures) 

 some of the electrons were caught by the bar connecting the 

 two disks. In the experimental arrangement described in 

 the present paper this possible source of error is absent. 

 Moreover, the electromagnet used in these later experiments 

 produced a strong uniform magnetic field over a space much 

 larger than that occupied by the wire and the electrode, and 

 was in this respect greatly superior to that at disposal in the 

 earlier work. These considerations probably account for 

 the fact that in the earlier experiments some electrons 

 succeeded in reaching the electrodes in spite of the magnetic 

 field. 



Summary. 



The nature of the carriers, in a vacuum, of the negative 

 thermionic currents from palladium, pure and commercial 

 platinum, and iridium has been studied. It is found that 

 the carriers consist almost entirely of free electrons. If any 

 of the discharge is carried by heavy ions, then the pro- 

 portion of heavy ions to electrons is, for all temperatures, 

 certainly less than one part in two thousand, and probably 

 less than one part in ten thousand. 



Holt Physics Laboratory, 



University of Liverpool. 

 March 14, 1913. 



