' 
. 
Solar Spectrum caused by the Sun’s rotation. 823 
circle. 
Between the two D lines, the spectra of the sixth and eighth 
orders usually showed no less than eight other lines, most of 
which are supposed to be water lines, produced by the vapor 
In our atmosphere, and therefore of course not subject to dis- 
placement by the sun’s rotation. I was in hopes to make use 
em as reference points, and to determine the displacement 
of the D lines by simply measuring the intervals with the 
micrometer. I soon found, however, that the atmospheric lines 
Were too f 
bisection, 
the microm 
In a feeble light. I was accordingly compelled to make the ob- 
i] 
ee 
i= 
72 
oS 
fo 
cr 
=) 
ct 
o 
o 
S 
fe) 
=) 
os 
<3 
~ 
5 
fa) 
° . 
eh: 
ot 
cr 
ia] 
wn 
= 
5. 
i) 
8 
En) 
oC 
Qu 
oo 
4°) 
ie) 
9 
ee 
5 
? 
intervals This will account for the small variations of the measu 
prone between the same lines as determined on different occasions—variations 
derably exceeding the probable errors of reading. 
