Discharge in Rarefied Gases. 189 



to the equatorial plane, the deformations of the bright centre 

 bundle show the action of the magnet on those rajs which 

 make larger angles with the equatorial plane. Such an 

 oblique bundle rolls itself, when magnetized, into a helix whose 

 coils are steeper the greater the angle made by the rays with 

 the equatorial plane, and the closer the nearer they lie to the 

 magnetic pole. 



With increasing strength of magnetism, the coils of these 

 helices, of which the before-mentioned spiral forms a special 

 case, place themselves more closely about the magnetic curve 

 which passes through the point of origin of the rays, and 

 change into it as far as the eye can see. Strictly, then, 

 the magnetic curve is only the geometrical axis of the true 

 form of magnetized light. 



We see, from what has been advanced, that the forms of 

 the magnetized rays are those which a linear conductor, free 

 at one end and fixed at the other, would take up, if it, endowed 

 with a certain amount of rigidity, were traversed by a simi- 

 larly directed current, similarly placed with reference to the 

 magnet. 



If, now, the magnet acted upon a conductor composed of as 

 many pieces arranged in line along the direction of the currents, 

 which pieces jvere all fixed at the end turned towards the 

 negative pole, or at least displaced with difficulty in a direction 

 at right angles to the direction of the current, but free at the 

 other end, such a system, by breaking up into as many 

 separate magnetic curves as there are separate currents, 

 would show precisely the phenomena which the stratified dis~ 

 charge presents to the magnet. The phenomena, on the con- 

 trary, would be impossible if all the layers together formed 

 only a single current between the kathode and anode. 



The independence of the different portions of the discharge, 

 for example, of the kathode-light Avith the first layer of the 

 positive light is seen from the following. When the kathode- 

 rays unroll themselves spirally, the first layer of the positive 

 light does not follow the end of the negative rays in its revo- 

 lutions, but the layer remains on the outside of the whole 

 spiral on the side turned towards the anode, without having 

 any connexion with the end of the ray on the interior of the 

 spiral. Each layer behaves in the same way towards the pre- 

 ceding layer of positive light on the negative side. 



In conclusion, it is easy to see how the view above explained 

 removes the difficulties which arise out of the view till now 

 usually held. From the kathode, as from a number of points 

 lying between the two electrodes, which correspond to 

 the limits of the positive layer towards the kathode, issue 



