electro-magnetic fields on the spark spectra of gold, etc. 83 



ammeter. The magnet was arranged broadside on, that is, the 

 lines of force between the poles were in a direction perpendicular 

 to the line joining the slit and the grating. 



The ends of the electrodes were held firmly in the middle of 

 the gap between the poles by means of a specially constructed 

 mechanical arrangement. The stand of this holder is firmly 

 clamped to the stage of the magnet, and it has three movements. 

 The holder can be moved upwards or downwards; also in a direc- 

 tion perpendicular to this; and thirdly there is a slow adjustment 

 to keep the ends of the electrodes at a constant distance from 

 each other. This slow adjustment is very necessary, because, as 

 the ends of the electrodes wear away during the sparking, and 

 the distance between them becomes greater, the spark begins to 

 discharge to the poles of the magnet. The distance between the 

 ends of the electrodes was usually from 1 to 2 mm. 



The' electrodes of the metals gold, tin, and lead were thin 

 strips of about 2 mm. wide and 1 mm. thick. Those of antimony 

 and bismuth were made by melting the metals, and sucking the 

 molten metal into narrow glass tubes of about 2 mm. internal 

 diameter. The glass and its solidified contents were then thrown 

 into water, and it was easy to pick off the shattered glass from 

 the surface of the pencil of metal. The pieces were then carefully 

 filed flatter. 



The spark was focussed on the slit by a quartz lens of a focal 

 length of about 5 inches and of 1 inch diameter. It was firmly 

 fixed in a holder, and it could be moved upwards or downwards 

 by a fine screw through a distance of 2 or 3 mm. By this means 

 two photographs could be taken on the same plate, one parallel 

 to the other — one of the normal spark and the other when the 

 spark was discharging in the magnetic field. 



In order to analyse the nature of the vibrations in the magnetic 

 field, a calcite crystal was introduced between the quartz lens and 

 the slit of the spectroscope. The crystal was specially polished 

 at the rhombic ends, and it was selected of a sufficient size to 

 disperse the two images of the ordinary and extraordinary rays 

 at a convenient distance from each other. The two images were 

 about 25 mm. from and above each other when they fell on the 

 slit ; and the optic axis of the crystal was perpendicular to the 

 lines of force between the poles of the magnet. By this arrange- 

 ment the "ordinary" and the "extraordinary" vibrations of the 

 same line could be photographed on the same plate. 



As a rule, any division of the lines into triplets is well marked 

 when the lines are observed in the second order. As regards the 

 lines observed in the first order the division is not so well marked, 

 unless they are sharp and narrow. When any line is normally 

 strong and broad, a widening may be noticed, and it is then 



