on the spark spectra of titanium, chromium and manganese. 77 
X 
dX / X 2 
2666-12 
+ 2-12 
0 
- 2-12 
2663-45 
+ 2-10 
0 
- 2-07 
2653-63 
+ 2-00 
0 
- 2-01 
The lines in the ultra-violet beyond 2653 - 63 were very weak, 
and there was very little to be seen on the photographic plate. 
Several of them were noted, but they were too weak to measure ; 
such are : — 
2608-01 
2590-88 
2563-71 
2533-50 
2530-06 
There were also many other lines which were divided or widened, 
but they were too weak to measure. Only those lines are described 
which clearly affect the photographic plate during an exposure of 
30 minutes. 
The lines 399T31 and 2953'46 were neither widened nor divided ; 
they appeared to become a little sharper. 
All too weak for analysis. 
The following were divided into two, but the constituents had 
not all the same appearance. In some of them the outside edges 
were more diffuse than the inside edges, and it is probable they 
were quadruplets not unlike those described on p. 65, but they 
were too weak to measure on analysis by the calcite. In others, 
there did not appear to be any difference in the diffuseness 
of the edges ; but they also may be aggregates of several con- 
stituents. The measurements were taken between the centres of 
the aggregates. 
3991-31 
3976-82 
3969-93 
+ 0-47 
0 
- 0-47 
+ 2-07 
0 
- 2-07 
+ 0-91 
0 
- 0-91 
The outside edges of the constituents were more diffuse than 
the inside ones. 
-Too weak to analyse. 
