Perkins — Rectification Effect in a Vacuum Tube. 487 



A careful study of the critical points of the curve shows 

 certain indentations in the maxima that persisted in quite an 

 unaccountable manner, and were nearly similar in the two 

 maxima; also curious humps in the curve near the electrodes, 

 where a very small current could be produced. Fig. 4 gives 

 half of the curve exhibiting these effects. A further series 

 of readings was taken with a storage battery interposed in 

 the galvanometer circuit. A few volts only were needed to 

 reverse the current when the coil was near the center, and a 

 large current in the opposite direction was readily produced ; 

 but 100 volts or more were v needed at points near ends. The 

 current produced by a given battery E. M. F. while the excit- 

 ing ring was moved from A to B is given in fig. 5. It will 

 be noticed that at the center larger values were obtained when 

 the two pressures in the tube were opposed, than when they 

 acted together ; also that the ionization was more available 

 for carrying a current when the ring was near the center. 



In order to study the potential differences between A and B 

 for different settings of the ring, an electrostatic voltmeter of 

 the Kelvin multicellular type was used, and fig. 6 (p. 490) 

 shows how the voltage varied. This bears out the indications 

 of the battery effect mentioned above, where the pressure 

 needed to reverse the current was a maximum at* the ends. 

 It will be seen that the voltmeter indicated the largest values 

 when the ring was near the terminals, although the limited 

 range of the instrument prohibited a complete set of readings. 

 It was also most significant to note that the time required for 

 the "needles" to come to rest varied with the magnitude of 

 the deviation. The small values near the center were reached 

 in a fraction of a second ; while for the large readings near 

 the ends it required several minutes to fully charge the volt- 

 meter to the indicated potential. This again bears out the 

 evidence of the other curves that indicate a maximum current 

 effect when the ring is near the center of the tube. 



Any theory accounting for this phenomenon must explain 

 the following facts : 



(1) A direct current caused by an alternating electrostatic 

 field. 



(2) The positive pole of the tube is always the one nearest 

 the exciting ring. 



(3) The maximum currents are obtained at points near by, 

 and on either side of the center. 



(4) The terminal voltage of the tube varies oppositely from 

 the current ; it is minimum when the current is maximum. 



JSTo attempt will be made to account for the indentations 

 already referred to, and other minor effects ; they may be due 

 to minute disturbances, the shape of the tube, terminals, etc., 

 and so will be ignored. 



