1882.] on Matter and Magneto-Electric Action, 81 



successive discharges (in tlie same direction) with a weak magnetic field 

 and a slowly revolving mirror ; Fig. 4 the same, with a slightly more 

 rapid rate of revolution ; Fig. 5 a single discharge, with a stronger 

 field and greater speed of mirror ; Fig. 6 a single discharge in a 

 strong field, with a still greater speed of mirror. It should be men- 

 tioned that in all these figures the images to the left are to be 

 regarded as anterior to those on the right, and that they represent 

 various phases of the left-hand discharge in Fig. 2. 



If, however, we observe the right-hand discharges with a mirror 

 revolving in the same direction as before, it is clear that the actual 

 curvature of the discharge will be turned in the opposite direction 

 (with reference to the motion of the mirror) to that in the case of the 

 left-hand discharges. The consequence will be that the appearance 

 in the mirror, when the rate of revolution is not too great, will be 

 something like Fig. 7, instead of Fig. 6. As the speed of the mirror 

 is increased, the convexity will diminish, and ultimately be replaced 

 by a concavity of the same kind, although not so marked, as that in 

 the case of the left-hand discharges. 



These diagrams show that each coil discharge commences with a 

 bright spark passing directly between the terminals ; that this spark 

 is in general followed by the pink light or arc discharge, which passes 

 first in the immediate neighbourhood of the initial spark, and gradually 

 extends like an elastic string in semicircular loops outwards; and 

 that the flame proper is a phenomenon attendant on the close of the 

 entire discharge. It should be added that observations with a mirror 

 revolving on a horizontal axis, and with a horizontal slit in front of 

 the discharge, show that the disk is not simultaneously illuminated 

 throughout, but that it is a locus of a curvilinear discharge which moves 

 outwards and expands in its dimensions from the centre. 



The mechanism of the discharge would therefore seem to be as 

 follows : In the first place, as soon as the tension is sufficient, the • 

 electricity from the terminals breaks through the intervening air, 

 but with such rapidity that the fracture is like that of glass, or other 

 rigid substance. This opens a path, along which, if there remains 

 sufficient electricity of sufficient tension, the discharge will continue 

 to flow. During such continuance the gas becomes heated, and 

 behaves like a conductor carrying a current ; and upon this the 

 magnet can act according to known laws. As long as the electricity 

 continues to flow, the heat will at each moment determine the easiest, 

 although not the shortest path for its subsequent passage. In this 

 way the gas, which acts at one moment as the conductor of the dis- 

 charge, and at the next as the path for it, will be carried further and 

 further out until the supply of the electricity from the coil fails, and 

 the whole discharge ceases. We are, in fact, led by these experiments 

 to the conclusion that it is the gas in the act of carrying the current, 

 and not the current moving freely in the gaseous space, upon which 

 the magnet acts. 



This explanation of the magnetic displacement of a discharge 

 Vol. X. (No. 75.) g 



