44:2 Mr. J. Henry on the Deflexion by Magnetic 



mirror. The magnetic force deflected this discharge at first, 

 bat immediately it changed its appearance it fell back to the 

 straight line between the electrodes. The change in the 

 character of the discharge was evidently due to the increase 

 of pressure in the discharge-tube caused by the heating of 

 part of the gas, as letting more gas into the tube would have 

 produced the same changes. This experiment was repeated 

 several times, but always with the same result : after stopping 

 the discharge for a few seconds, it always started again as 

 at first. 



In hydrogen the deflexion is very small, but as in other 

 gases it increases, though slowly, as the pressure of the gas 

 falls. For the lowest pressure obtained — less than one millim. 

 of mercury — the deflexion was very much greater than for 

 pressures of one or two centims., yet it was not nearly so great 

 as that obtained in air or carbonic acid gas for the same 

 pressure. 



The discharge in hydrogen, when examined in the rotating 

 mirror, showed no appreciable broadening until the pressure 

 of the gas in the discharge-tube was reduced sufficiently to 

 allow a decided magnetic deflexion, and as the pressure was 

 lowered still furth r, the broadening of the discharge in the 

 mirror, as well as the magnetic deflexion, continued to increase. 



Explanation of Results. 



These experiments point to the conclusion that the influence 

 of the gas on the deflexion of the discharge in it is indirect. 

 In any of the gases used, when precautions were taken to make 

 the discharge abrupt, and to prevent it trailing out at the 

 end, no deflexion could be observed, even in a very strong 

 magnetic field produced by a large electromagnet with 

 conical pole-pieces, the discharge-tuhe used in this case being 

 small — one inch in diameter — to allow the pole-pieces to be 

 brought near together so as to increase the magnetic force 

 between them. But when the discharge was allowed to pass 

 in the ordinary way, the deflexion always increased as the 

 discharge became less abrupt and more prolonged as the pres- 

 sure of the gas in the discharge-tube fell. 



A very important factor in determining the magnitude of 

 the deflexion in any gas for a given frequency of discharge, 

 was its duration, the quantity of electricity carried by the dis- 

 charge being of little importance. 



This was shown by a comparison of the deflexions obtained 

 in the experiments, case (4) above ; when the pressure of the 

 gas in the discharge-tube was high, and a small leyden-jar 

 was used instead of the two shown in fig. 13, the quantity of 



