OF SPECTPwi AT THE SOLAR ECLIPSE OF JANUARY 22, 1898. 
393 
intensity of the corresponding dark lines is given in column 3. Although the 
intensities of the bright lines, estimated on a scale of 1 to 10, are not comparable 
with those of Rowland ranging from 1 to 1000, the relation between the bright and 
dark line spectra is, nevertheless, clearly indicated by the regular progression of the 
figures in the last column of both tables corresponding with the progression between 
0 and 5 in the middle column :—■ 
Intensities of Iron and Titanium Lines in the Chromosphere and in the Solar 
Spectrum. 
Number of lines in 
flash speetrum. 
17 
33 
25 
10 
Iron Lines. 
Photographic 
intensity. 
0 
1 
o 
o 
Average intensity of the corre¬ 
sponding dark lines in © (Rowr.AND) 
5 5 
7'5 
11 3 
25-3 
Total 85 Weighted mean 1'3 Weighted mean 10'2 
Number of lines in 
flash spectrum. 
5 
9 
13 
9 
4 4 
1 5 
Total 41 Weighted mean 2'0 
Average intensity of the corre¬ 
sponding dark lines in © (RowmND) 
1-G 
3- 4 
4- 2 
5- 4 
7-7 
10-0 
Weiobted mean 4’5 
Titanium Lines. 
Photographic 
intensity. 
0 
1 
2 
The mean intensities given at the bottom of each .column show the greater 
intensity of the Ti lines in the chromosj^here compared with Fe, and the much 
greater average intensity of the Fe dark lines in the solar spectrum compared with 
the Ti lines. 
Tlie striking dissimilarity in the relative intensities of the lines of different 
elements in the bright-line and dark-line spectra is j^robably due to the unequal 
heights to which the gases of the various elements ascend in the chromosphere. 
The intensities as they appear in photographs of the flash spectrum are evidently 
largely determined by the relative depths of the various gaseous strata, the more 
extensively diffused gases giving the strongest bright arcs simply by reason of the 
greater radiating area. The low-lying gases, on the other hand, although they may be 
VOL. CXCYII.—A. 3 E 
