•138 Mr. J. Henry on the Deflexion by Magnetic 



from any cause unless each discbarge lasted a sensible time 

 ('Recent Researches/ p. 105, par. 90). 



Support is given to this view by the fact observed throughout 

 the experiments, that when the discharge was vertical a hori- 

 zontal magnetic force always increased the potential-difference 

 between the electrodes of the discharge, this increase being- 

 greatest \n hen the deflexion of the discharge was greatest — 

 the mean potential-difference was increased in some cases 

 from 1000 to 1100 volts, — while in hydrogen, where the de- 

 flexion was hardly visible, the same magnetic force produced 

 no perceptible increase in the potential-difference, as mea- 

 sured by the electrostatic voltmeter used. When the discharge 

 was horizontal and at right angles to the magnetic force, the 

 potential-difference between the electrodes was increased or 

 diminished according as the magnetic force increased or 

 diminished the deflexion of the discharge due to the convection- 

 currents in the gas. In the latter case the magnetic force 

 shortened the length of the discharge, and at the same time 

 tended to keep the ionized gas together, thus reducing the 

 resistance, while it had the opposite effect in the former 

 case. 



When the time between successive discharges was increased 

 by using a hand-break for the primary current in the induction- 

 coil, the magnetic deflexion was diminished until, when the 

 discharge was separated by nearly a second or more, each 

 discharge began almost in the straight line between the 

 electrodes, and was broadened as shown in fig. 12, the ionized 

 gas being here deflected through the breadth of the band 

 while the discharge lasted. 



Fig-. 12. 



To get a visible broadening the pressure of the gas in the 

 discharge-tube had to be low enough to let the discharge last 

 some time : for high pressures of the gas the discharge passed 

 almost straight across. 



Farther Experiments. 



To try if the current in the discharge as measured by the 

 galvanometer was proportional to the maximum current 

 during the discharge, a magnetizing-spiral was connected in 

 series with the discharge-tube ; a bundle of steel wire carried 

 in the end of a glass tube could be slipped into this spiral and 



