206 MESSRS. A. SCHUSTER AND G. HEMSALKCH ON 



ordinary sparks at 4703 is also absent. When the spectrum is photographed on the 

 rotating film, the lines resolve into remarkable clouds (fig. 20, Plate 10), which do not 

 admit of measurement. The appearance of the lines shows, however, that the velocities 

 of the magnesium particles is great, and approximately the same as that of aluminium. 



Aluminium. The prominent lines are the two triplets beginning with 4512 (a) 

 and 3613 (/8) respectively, and the doublet which lies between H and K. The former 

 do not appear in the arc spectrum, which shows the violet doublet strongly. The 

 velocities we have obtained are the largest we have measured, but the displacements 

 (fig. 21, Plate 10) are very small, and the lines are measurable only near the poles, so 

 that the numbers are doubtful. The violet triplet seems displaced through a greater 

 distance than the blue, but the measurements are so uncertain that we have taken the 

 average displacement without distinguishing between them. The violet doublet almost 

 disappears in the rotating film, leaving only a faint cloudy formation similar to that 

 shown in the magnesium spectrum, but possessing peculiarities which require 

 further investigation. 



Manganese. Many lines of manganese appear on our stationary photographs, but 

 on the rotating film they completely disappear under the usual conditions of our 

 experiments. When the slit is widened and the speed reduced, the displacements 

 could be observed, but were not measured. 



Iron. The iron lines were not well marked, and completely disappeared in the 

 rotating films. 



Copper did not show any well-marked lines, the displacements of which could be 

 measured. 



Zinc. This metal has already been discussed in detail, and it only remains to 

 mention that the line at 4058 (e) disappears on the rotating film. 



Silver showed no prominent lines when our battery was discharged through it, but 

 the calcium lines are always seen (see p. 212). 



Cadmium. A number of cadmium lines appear on our photographs, which agree 

 in position with the lines so frequently measured ; but there is some difference in the 

 relative intensities as noted by different observers. The double line 5379 and 5339 (a) 

 shows a greater displacement than that of the other lines, and the distance between 

 the components is so great that the explanation given for the corresponding zinc 

 lines does not seem to hold here. Fig. 22 shows the behaviour of the cadmium lines. 



Mercury. Sparks were taken from a surface of liquid mercury, and. though the 

 displacements can be readily seen (fig. 23), and the edges of the lines are fairly sharp 

 at some distance from the electrode, they are so diffuse in close proximity to it that 

 no satisfactory measurements could be made at points corresponding to those taken 

 in the case of other metals. Better results were obtained when the electrodes were 

 either amalgamised zinc or cadmium ; but the measurements in these cases, though 

 they are given in the table, are difficult to make and cannot be trusted. When 

 cadmium amalgam is used (fig. 24), the photographs show clearly that the mercury 



