ELECTRON EMISSION FROM GLOWING SOLIDS. 
287 
possible effect of the electron emission from the tantalum wire itself. If the electrons 
emitted by the wire escape through the Nernst filament they would take part in 
carrying the measured thermionic current. In order to see whether the current so 
carried is likely to be a considerable fraction of the whole, a piece of tantalum wire of 
the same dimensions as that used inside the filament was fitted in the apparatus, and 
the emission from this was measured at different temperatures.' At the highest 
temperature recorded in the table the emission was only 9'7 x 10~ 7 amperes. It is, 
however, probable that in the experiment proper the wire was considerably hotter 
than the filament surrounding it, and there is therefore some doubt as to what part 
of the measured current may be due to the tantalum ; but I think it can safely be 
stated that the emissions from the filament alone with the two methods of heating 
are of the same order of magnitude, and there is certainly no such difference as would 
be expected if the emission were caused by the re-combination of the electrolytically 
separated elements of which the filament is composed. In order to eliminate the 
uncertainty as to the effect of the heating wire the following experiment was made. 
III. A Comparison of the Electronic Emission from a Nernst Filament Cheated in 
the Ordinary Manner with that given by the same Material when ELeated upon 
Platinum. 
In making this comparison the emission from a filament at different temperatures 
when heated by an alternating current in a good vacuum was investigated as before. 
When the thermionic current tested under similar conditions on different days had 
become fairly constant, the filament was detached from the apparatus and a piece of 
fine platinum tube was substituted for it. The external diameter of the platinum 
tube was a little less than that of the filament, and the internal diameters of the two 
were the same. The length of the tube used was rather greater than that of the 
glowing part of the Nernst filament so as to make the surface areas about the same 
in the two cases. A thermocouple of fine wires of platinum and platinum-rhodium 
was carefully welded on to middle of the platinum tube. This junction was 
standardised in the usual manner and served to determine the temperature of the 
tube during the experiments. Observations of the negative emission from this 
platinum tube were then made under similar conditions to those employed with the 
Nernst filament, except that the temperature was determined by the thermocouple 
and not with the pyrometer. The emission from this platinum tube was rather 
larger than I had expected it would be, and much larger than that measured from 
the tantalum wire. It was therefore taken down and boiled in strong nitric acid and 
afterwards heated for some time in air at atmospheric pressure. This treatment 
reduced the emission considerably, but the thermionic current at 1516° C. was found 
to be 9xl0~ 6 amperes per sq. cm. of heated surface, which is about 20 times the 
value to which the emission from a platinum wire can be reduced. However, this 
value of the current was fairly constant, and the values now obtained at different 
