256 



Dr. R. S- Willows on the Effect of a 



Fig. 5. 



potential gradient varied at every point of the tube when the 

 magnetic Held was on. 



For this purpose a tube was used like that employed by 

 H. A Wilson * for a similar determina- 

 tion without the magnetic field. Two 

 electrodes A, B were fastened together 

 at a fixed distance apart by means of three 

 thin glass rods. The bottom electrode B 

 was fixed to a narrow tube 0, closed at 

 its bottom end, and down the centre of 

 which passed a wire ; this made contact 

 with an. adjustable column of mercury on 

 which A, B, C floated. The current was 

 brought to the electrode A by means of 

 a thin spiral spring D, and passed out. 

 by the mercury column and a wire pushed 

 into this past the screw-clip M. This 

 last allowed of a fine adjustment of the 

 length of the mercury column. By this 

 means any part of the discharge could be 

 brought near two exploring electrodes N, 

 about 2 mm. apart, put in the tube by a 

 sealing-wax joint. These explorers were 

 connected to opposite pairs of quadrants 

 of an electrometer. The readings of the 

 latter would thus be proportional to the 

 electric force at the mean position of the 

 explorers. The distance between A, B 

 was about 7*5 cm. 



I have to thank Prof. Thomson for pro- 

 viding me with a means of producing 

 a uniform transverse magnetic field. 

 This consisted of an electromagnet built up out of a large 

 number of transformer rings with a gap cut in each. The 

 mean diameter of a ring was 18 cm., the width of the annulus 

 and of the gap was 3 cm. in each case. By this means a 

 fairly uniform field could be produced over a length of about 

 16 cm., so that the whole length of the discharge could be 

 kept in the field while it was being explored. The field 

 strength varied from 200 to 1000. Such a field was found 

 not to increase the current through the tube at low pressures. 

 It has already been shown that an increase in current is only 

 brought about when the field exists near the cathode : for all 

 other positions the current is lessened. When the field exists 

 along the whole length of the discharge, it appears from the 

 * Phil. xMa ? . June 1900. 



^ 



