ox A TABLE OF STANDARD WAVE LENGTHS OF THE Srr.CTRAL LINf^: 
117 
used tlie plates originally obtained for metallic spectin. generally ii>ing tlie lines dno 
to the impurities. The method, I believe, is obvious from the faltle. For ,. 1 
substandards are necessary, but are used whenever ihvy lieroiiK* ^n. 
(Hi J[ 
region no 
TABLE 1. 
PRELIMINAKY OBSERVATION? BY MR, KOYLE, 
Made at Baltimore in 1882 and 1883. 
Spectra. 
From 
5G24. 23 
and 
5C24.76. 
4215.67 
4325 02 
4376.06 
4497.06 
4501.44 
4691.64 
4824.34 
4919.15 
4920.70 
4924.13 
4991.39 
5006.31 
5060.22 
5109.82 
5110.58 
5162.51 
From 
4824.34. 
5353.64 
5361.83 
Table." ^* 
i 
Spectru. 
4 
From 
and 
5624.76. 
5624.23 
5624.7G 
6246,r)8 
6322.91 
6431.09 
6563.04 
6750.41 
6752.92 
4824 3*. 
6024.31 
GO 27 .31 
078«.ll 
5791.21 
5bUii.ll> 
5^03.07 
5^06.17 
5914.30 
r ..1 
' i 
4 I « 
O^ 
.uu 
f 
n-s4,n M^ 
• This column lias been added from the final table of standards, ^'^"^^^^^^^^^^ The difTeri 
on account of the inferior grating he used. The comparison \s tims o i 
to want of precautions which were finally adopted. 
» are prob«iW> du 
VOL. ir. 
10 
