123 Prof. Pliicker on the Action of the Magnet 



attains a length of 80 millims. This dark space separates 

 sharply the stratified light we have been hitherto considering, 

 from the light which appears about the negative electrode (the 

 warmtli-pole) itself, and which forms a kind of atmosphere 

 around it. This remarkable light will occupy our special atten- 

 tion at the end of this paper. 



8. Finally, the negative electrode is, in its immediate neigh- 

 bourhood, surrounded by an envelope of variously coloured, 

 finely stratified light, which corresponds with the form of the 

 electrode. The fine layers are spherical when the positive elec- 

 trode itself has a spherical form ; they have a cylindrical form 

 in the case where the electrode projects into the tube as a 

 wii'e. 



9. After these preliminary observations, I proceed at once to 

 the proper subject of this paper, namely, the action of the mag- 

 net upon the electrical discharge through spaces containing 

 rarefied gases. In the experiments immediately to be described, 

 I employed a great upright horseshoe magnet, to the two limbs 

 of which two heavy armatures were applied, 4 centims. thick, 

 13 centims. wide, and 20 centims. long. Each of these arma- 

 tures was rounded circularly at one end, and the rounded ex- 

 tremities were directed towards one another, being kept, by an 

 intei'posed brass disc, at a distance of about 4 millims. The 

 Geissler tubes were placed upon these armatures, first, in an 

 equatorial direction ; secondly, in a position perpendicular to it, 

 that is, in an axial direction. 



10. If, instead of the tube, a freely moveable copper wire, 

 through Avhich a galvanic current passes, be brought in the 

 (horizontally) equatorial position, it will be either drawn vertically 

 downwards or repelled upwards, according as the direction of the 

 galvanic current in the wire corresponds, or the reverse, with the 

 superior directions of the Ampereian currents, which are the 

 same in direction in the two approximated armatures, and may 

 be supposed to move in vertical planes perpendicular to the 

 axial direction. (If we represent the direction of the Ampereian 

 currents by the path of the hand of a watch, we must of course 

 regard the watch itself as a magnet whose south pole is the 

 lettered watch-plate.) The attraction or repulsion of the con- 

 ducting wire may also be regarded as the resultant of two couples 

 which tend to turn the wire in opposite directions. 



. Further, if instead of the tube we imagine the conducting 

 wire to be placed in an axial direction, in such a manner as to 

 extend equally far on both sides beyond the point of maximum 

 action, tlien the two halves of the conducting wire are subject to 

 opposite actions of the magnet. Supposing the two extreme ends 

 to be rigidly fixed, the two halves would strive to revolve in a 



