upon the Electrical Discharge in Rarefied Gases. 123 



horizontal plane in opposite directions around the two ends, as 

 though endeavouring to break the wire in the middle (PI. I. fig. 2). 



11. I placed a tube about 270 uiillims. long, widened in the 

 middle to an ellipsoid (width of tube about 12 millims., greatest 

 width in the middle 52 millims.), equatorially upon the ap- 

 proximated armatures. This tube contained a trace of phos- 

 fdiorus, and gave a beautiful red light when the discharge wag 

 ed through it by means of the two platinum wires fused into its 

 extremities. In the tube itself the sharply-defined stratification 

 of the light appeared ; on entering the ellipsoid the strata spi'ead 

 out in a spherical form, becoming thereby more strongly bent j 

 and having become plane in the middle of the ellipsoid, they 

 changed the direction of their curvature. 



According to the direction of the current and to the manner 

 of exciting the magnet, both of which may be reversed, four 

 distinct cases present themselves; these, however, require no 

 special attention, inasmuch as they are the same as in the case of 

 the conducting wire through which a galvanic current passes. 

 Jn two cases the electrical light-currents were attracted in the 

 ellipsoid; in the other two they were repelled. In the case of 

 attraction (fig. 3), the light-current declined from the side of the 

 positive electrode (the light pole) into the ellipsoid, and termi.- 

 nated immediately above the approximated armatures in a gra- 

 dually-narrowing sharply-defined point of continually increasing 

 brightness. From the other side, beautiful red undulatory 

 flames proceeded, which sunk down in the ellipsoid, and ex- 

 tended beyond the middle point between the armatures, without 

 manifesting any tendency to unite with the steady light-stream 

 from the opposite side. On changing the direction of the mag- 

 netic current (fig. 4), the phsenomenon underwent a gradual 

 alteration. On entering the ellipsoid the light became concen- 

 trated into a luminous arch, which traversed the upper part of 

 the bulb in the equatorial plane. Immediately on entering the 

 ellipsoid the stratification of the light became finer; at the upper 

 part of the arch, where the concentration of the light was great- 

 est, the dark intervals became more and more numerous and 

 distinct. A simultaneous change in the direction both of the 

 electric and magnetic cui-rents, eff'ected, as was of course to be 

 expected, a mere inversion of the original phsenomenon, the 

 light-stream sinking to a point from the opposite side. In the 

 second phaenomenon (that of repulsion) a similar double inversion 

 of the direction of the currents efl"ected no appreciable alteration. 

 The repelled arch of light did not undergo any apparent change 

 on changing tbe direction in which the positive electric current 

 entered the ellipsoid, which was clearly owing to the greater 

 distance of tbe arch from the poles. 



