148 Electrical Discharge Phenomena in Rarefied Gases. 



Two or more rings could be made to appear by placing an electri- 

 fied platinum circle of wire equatorially within the bulb. When the 

 platinum circle was negatively electrified, the luminous ring was 

 repelled by it. In this manner the ring itself was invariably shown 

 to be negatively electrified. Its direction of rotation was found to be 

 that of the current induced in a loop of wire when the loop is suddenly 

 moved up to a north magnetic pole — clockwise, looking through the 

 loop at the pole. The outside of the glass bulb was always negatively 

 electrified when a luminous ring appeared in the interior. This 

 pointed to the removal of a layer of positively electrified gas from 

 the inner surface of the bulb through the action of the magnetic field. 

 Although such radial streams of positive ions so produced might 

 account for the luminosity of the ring through their collisions with an 

 accumulation of negative ions at the more central part of the bulb, 

 they would not have produced rotation of the luminous ring in the 

 direction already observed. The incoming radial streams of positive 

 ions were studied in detail with an apparatus more suitable for 

 examining the clis electrifying action of the magnetic field. Those 

 experiments established two facts, viz., that the loss of positive electri- 

 fication from charged bodies is brought about by the magnet, through 

 the concentration of negative ions which occurs at the strongest part 

 of the magnetic field immediately the electrodes are magnetised, and 

 also that the luminosity of the ring itself is due largely to the collisions 

 between the incoming streams of positive ions and this accumulation 

 of negatively electrified gas between the pointed ends of the electrodes. 

 A potential difference is thus set up within the bulb between the 

 negative gas-mass at the centre and the positively electrified layer of 

 ions residing upon the inner surface of the glass, which rapidly reaches 

 a value sufficient to give rise to a discharge through the residual gas. 

 It is then that the positive ions stream inwards, accompanied by a 

 corresponding outward-moving whirl of negative ions. 



Experiments upon the effect of causing the magnetic field to either 

 slowly or rapidly reach its maximum value, as well as diminish either 

 slowly or rapidly to zero, have shown that the rate of change of the 

 magnetic lines plays an important part in the actions here described. 

 A very rapidly growing field would diselectrify a positively charged 

 body, whereas, when the magnets were slowly increased in strength 

 there was no diselectrification in such cases. In certain experiments, 

 the act of suddenly destroying the magnetic field produced diselectri- 

 fication, while if the current were slowly diminished in the coils of the 

 electro-magnets there was no evidence of any such effect. 



Both the luminous ring and the diselectrification phenomena are 

 attributable to the same causes. The direction of rotation of the ring, 

 however, forms a difficulty, on the assumption that a rapidly moving- 

 ion is equivalent to a current along a flexible conductor. Incoming 



