218 MM. A. de la Rive and E. Sarasin on the Action of 



manometer, which during the whole experiment had remained 

 in communication with the glass tube. 



After repeating these experiments a few times, one can make 

 them with great rapidity and yet obtain, for the same gas in the 

 same conditions, almost absolutely concordant results. 



We executed several series of comparative experiments with 

 air, hydrogen, and carbonic acid, at pressures of 20, 30, and 40 

 millims. We limit ourselves to giving the results of some ; all, 

 however, were perfectly concordant with one another. 



Atmospheric air. Hydrogen. 



Defl. Pressure. Defl. Pressure* 



of galv. millims. of galv. millims. 



30° without magnetizing 20 30° without magnetizing 20 



with magnetization 6*64 with magnetization 3*50 



6-56 2-90 



6-20 3^0 



6 ' 58 Mean . . 3-20 



Mean . . 6*50 



From this Table is seen how the influence of magnetism on 

 the electric conductivity of a gas traversed by the induction- 

 current varies from one gas to the other. The numbers just 

 given show that magnetism diminishes the electric conductivity 

 of hydrogen much more than that of air; yet the apparent con- 

 traction of the jet under the action of magnetism is much less in 

 hydrogen than in air. This at once makes it evident that the 

 condensation of the jet is not the sole cause of the augmentation 

 of resistance, and that probably it is not even the preponderant 

 cause. 



The following are the results obtained in a series chosen from 

 several others which we made on the three gases in question : — 



Initial pressure of the three gases . . 30 millims. 

 Constant deflection of the galvanometer. 30° 



Pressure under the action of magnetization : — 



Hydrogen. Carbonic acid. Air. 



millims. millims. millims. 



7-30 10-60 13-70 



7-60 10-42 13-34 



7-60 1055 — 



Mean . 750 10-52 13-52 



The same result at 20 millims : 



Initial pressure of the three gases . . 20 millims. 

 Constant deflection of the galvanometer. 30° 



