1882.] on flatter and Magneto-Electric Action. 83 



the air; and, that done, it ceases. Hence, neither a magnet, nor a 

 blast of air will have any effect in diverting such a discharge. 



As a last stage of the phenomena, it may be mentioned that, if the 

 interval between the terminals be near the limit of striking distance, 

 either a blast of air, or the setting up of a magnetic field will alike 

 extinguish the discharge. 



Our experiments have been thus far carried on in air at atmo- 

 spheric pressure; but there is nothing in this pressure which is 

 essential to them or to the conclusions to which we have been led. 

 We may therefore repeat them in air, or any other gaseous medium, at 

 any pressure we please. This consideration leads us into the region 

 (so fertile in an experimental point of view) of discharges in vacuum 

 tubes. 



Commencing with a tube of moderate diameter and of very slight 

 exhaustion, we can at once recognize our former phenomena slightly 

 changed. Proceeding to another tube, of larger diameter and of 

 moderate exhaustion, and placing it axially or equatorially in a mag- 

 netic field, we see not only that the discharge (or rather the con- 

 ductor carrying it) is displaced, but also that the displaced ' part is 

 spread out into a sheet or ribbon, showing that the discharge is 

 affected gradually, exactly in the same way as was found in the 

 open air. 



When the exhaustion is carried further, the phenomena become 

 rather more complicated. At an early stage there is a distinct separa- 

 tion between the "negative glow" and the rest of the luminous 

 column ; and at a more advanced stage the column itself is broken into 

 separate luminosities or striaG. When this is the case, it is usually 

 said that the negative glow follows the lines of magnetic force, while 

 the luminous column distributes itself according to Ampere's law. 



It will, however, be found that when completely analysed the 

 action of the magnet upon the stride, taken individually, is the same 

 as that upon the negative glow, due allowance being made for the 

 differences in local circumstances subsisting between the one and the 

 other. We have elsewhere shown that the negative glow is in reality 

 as truly a stria as any other individual member of the luminous 

 column ; but with this difference, that it is anchored to, and dependent 

 for its form on, a rigid metallic terminal, whereas each of the others 

 is dependent on the variable form and position of the stria immediately 

 next in order, reckoning from the negative end of the tube. The 

 action of a magnet in throwing the negative glow into a sheet of 

 light, which is the locus of the lines of force passing through the 

 terminal, and which consequently varies with the position of the tube 

 in the field, is a phenomenon so well known that we need repeat only 

 a single experiment by way of reminder. 



Although it is not altogether so easy to show that the other stri^ 

 are directly affected by a magnetic field in the same way as is the 

 anchored stria, we may still satisfy ourselves that it is the fact, from 

 the consideration that when the stride are well developed and the 



G 2 



