TROWBRIDGE. — ELECTRIC DISCHARGES. 633 



made the pole of a strong electro-magnet. This retreat, while the mag- 

 netic field remains constant, occupies several minutes and seems to indi- 

 cate a decrease in the conductivity of the gas due perhaps to an increase 

 of temperature caused primarily by an increased admittance due to the 

 magnetic field. 



II. Magnetic Field and Electric Discharges. 



I wish to acknowledge my indebtedness to previous workers in the 

 field which I enter, especially to Lehmann (Drude's Ann., No. 7, p. 1, 

 1902) ; and since Professor J. J. Thomson has given a valuable resume 

 of papers on this subject in his treatise entitled "Conduction of Electricity 

 through Gases, 1903," I refrain from repeating a bibliography here. 



The production of the phenomena described in this paper depends 

 primarily on strong, steady currents, high voltage, and powerful mag- 

 netic fields. The effects were excited in cylindrical glass tubes of large 

 diameter ; these tubes were always connected to the pump. The cur- 

 rent was regulated by a running water resistance and the amperemeter 

 and electrometer were suitably connected. 



The cathode consisted of a rod of soft iron 16 cm. in length, 2 cm. in 

 diameter. The rod was sealed into the tube by means of a platinum 

 wire which, in turn, was soldered to the rod of iron. This rod formed 

 the core of an electro-magnet. A similar rod of iron with a coil of wire 

 forming another equal electro-magnet could be placed in any position out- 

 side the tube. Thus the magnetic lines of force could be formed along 

 the line of flow of the current in the tube or at right angles to this flow. 

 It was realized early in the course of the investigation that the size of 

 the tube, the character of the electrodes, shape, and material, together 

 with the strength of the current, electromotive force, and strength of 

 magnetic field, could modify the phenomena in a perplexing way. Each 

 trial seemed to bring out a new phenomenon. I will select, however, 

 those phenomena which could be definitely repeated under known con- 

 ditions, and since mere descriptions without photographs of such phe- 

 nomena are not particularly clear or convincing, I have made a number 

 of photographs of certain typical phenomena. 



The mathematical theory of the effect of a magnetic field on electric 

 discharges in rarefied gases assumes, first, that if the viscosity of the gas 

 causes the velocity of the ion to be proportional to the force arising from 

 the electric intensity, the motion of the ion depends on the product of 

 the electric intensity and the strength of the field. According to the 

 vol. xli. — 42 



