on the spark spectra of titanium, chromium and manganese. 79 
2867-70 
d \/\ 2 
+ 2-36 
0 
-2-36 
1 The two constituents were equally sharp and strong. They 
f were very similar to those of 3336-49. 
2668-78 
2658-65 
+ 1-28 
0 
- 1-28 
+ 1-25 
0 
-1-25 
The constituents of these appeared to be equally sharp on both 
sides. 
Manganese. 
The lines were identified by reference to the measurements of 
Exner and Haschek (Sitz. d. k. Akad. Wien, vol. cv. n A . Abth. p. 399, 
and vol. civ. n A . Abth. p. 921). The constituents of the divided 
lines were much more diffuse than in other metals. The rapid 
oxidation of the metal also caused the spark to be irregular and 
this may have increased the diffuseness. 
The following lines were divided into four constituents, but the 
outside constituents of 3488'8 and 3482’9 were too weak and diffuse 
to measure. It will be noticed that the constituents of 3473'9 
are simple multiples of each other. 
3488-8 
+ 1-65 p 
0 
-l-65p 
? s 
3482-9 
? s 5 
4-0-87 p 
0 
-0-87p 
? s 
The two outside constituents were very weak and diffuse and 
impinge upon the two inside strong and sharp ones. 
3473-9 
4 - 3-13 s 
-1-51 p 
0 
- l - 51p 
-313 s 
The most refrangible constituent was weaker than the other 
- three ; and the two middle ones were the strongest. They 
were all very diffuse. 
The following liues were divided into triplets. The middle 
constituent was usually the strongest, and in some instances the 
two outside constituents were exceedingly weak as compared with 
the strong intense middle one. The middle constituent vibrated 
parallel, and the two outside ones perpendicular to the lines 
of force. 
