TATE AND FOOTE : ANODE FALL OF POTENTIAL 485 



ing of the voltmeter V. In the voltmeter method the relation 

 was obtained between the variable resistance R s and the volt- 

 meter reading. If a film of resistance R v exists at the probe the 

 relation between the observed voltmeter reading V and the 

 voltage drop on open circuit, V, is as follows: 



ti- T77 -tlr *$>r *l p 



Rv 



From this equation R P , the film resistance, is readily obtained. 



The values observed are plotted in figure 2. It appears 

 that the film resistance decreases with increasing current 

 through the probe for the range of currents used. Since the 

 probe acts as an anode during voltmeter readings evidence has 

 been obtained for the existence of an anode polarization similar 

 to the polarization at the cathode in a highly ionized gas, as 

 observed by Skinner. The present experiment would not 

 distinguish between a true polarization and a secondary anode 

 fall at the probe. This latter might be developed by the volt- 

 meter current and, although the current is very small, might 

 amount to several volts since, as is well known, the ordinary 

 anode fall is pronounced even for minute currents. However, 

 as Skinner has remarked, it is scarcely conceivable that a poten- 

 tial gradient could exist in a highly ionized space. The con- 

 ditions relating to the true anode and the secondary anode, or 

 probe, are not similar, since the ionization in the space surround- 

 ing the probe must be due almost entirely to the main current. 



The proof that the first explanation of the difference in volt- 

 meter and electrometer or potentiometer readings can not be 

 correct may be had in the following experiment. 



The temperature of the anode is a measure of the power 

 expended in it. The anode was made in the form of a crucible 

 and filled with tin — a very nonvolatile metal. The temper- 

 ature of the molten tin was measured by a minute rare-metal 

 thermocouple. A series of measurements showed that the tem- 

 perature of the anode (and hence the power dissipated there) 

 remained the same whether the voltmeter or potentiometer was 

 connected to the probe, although the power measurements by 



