126 frof. VlvLcker on the Jction of the Magnet 



coaoidal pole-point was screwed in. That bulb of the tube 

 which contained the warmth-pole was then so disposed upon 

 the conoidal poles, that the tube passing between these points 

 was situated in a horizontally equatorial position. Previous to 

 the excitation of the magnet, diffuse violet light was spread 

 through the bulb, surrounded by a pale green light, which ap- 

 peared to form a thin coating immediately upon the surface of 

 the glass bulb. After the magnet was excited, the diffuse violet 

 light collected to a horizontal, semilune-shaped, bright and 

 uniformly luminous disc, bounded towards the tube by an 

 almost circular well-defined concave arch, whose middle passed 

 through the point of the platinum electrode. This disc was 

 upon the opposite side enveloped by a narrow strip of beautiful 

 bright green light, which followed the cui'vature of the glass. 

 Moreover, the light (according to what has been already de- 

 scribed), which was red at a distance from the pole, but became 

 bright violet towards the pole, and showed the dark intervals 

 beautifully, descended towards the bottom of the bulb on 

 entering it from the tube, and terminated in a point. On 

 inverting the polarity of the magnet, the ph^enomenon remained 

 entirely unaltered, the tongue in which the light-stream termi- 

 nated merely passing towards the top instead of the bottom of 

 the tube — an alteration which we need no longer consider. 



21. On so changing the position of the tube (fig. 6) that, 

 while still remaining in a horizontally equatorial position, it no 

 longer passed between the polar points which continued in con- 

 tact with the bulb, the phsenomenon remained unaltered, with 

 the single exception that the horizontal violet disc, separated 

 from the glass by a green strip, was bounded towards the tube 

 by an arch whose convexity was now turned towards the tube, 

 and whose middle point still coincided with the point of the 

 platinum electrode. 



22. The two armatures were then moved apart from one 

 another, just sufficiently to allow^ the bulb to fit in between 

 them (fig. 7). The tube having the same direction as before, 

 the line joining the pole-points passed through the central point 

 of the bulb, where also the electrode terminated. On exciting 

 the magnet, the disc, which was formed as in the former cases, 

 was bounded by the axial diameter : it was, as before, so far as 

 the glass sides extended, enveloped by a beautiful green strip. 

 Upon the disc, in the neighbourhood of the axial diameter, a 

 strong accumulation of light took place, a phsenomenon which 

 was not observed in the previous cases. 



In the two last cases, as in the first, on inverting the mag- 

 netic polarity, exactly the same appearance was seen after some 

 time, as far at least as the disc was concerned. 



