140 
ON A TABLE OF STANDARD WAVE LENGTHS OF THE SPECTRAL LINES. 
From the foregoing we also find the mean of ^ to be 7621.282, and the absolute 
standard 7621.278. 
DETER3IINAT10N OF A. 
By 
the inicrometer across f 
the fi 
to th 
second spectrum of 
a grating witli 21 J- feet focus and lO^OOO lines to the inch, I obtained the wave 
without 
f lines in A more accurately 
he accuracy of the rest of tlj 
had ever before been obtained, b 
Tl 
of this was the very 
long distance to be measured in an interval 
y 
faint in light as to 
d 
necessary to open the slit wide. The following results were obtained : 
7594.07 
7G21.30 
7G23.57 
7G'24.86 
7593.98 
.15 
.38 
.71 
INIean. 
7594.02 
.22 
.47 
.78 
A 10,000 grating was finally found 
hich the A line 
uld be seen in 
the 
d spec 
hence its 
den 
th 
third spectrum d 
ed 
B 
the slit had to be so wide as greath 
mj 
green of the third spectrum was so bright as partlj 
the defi 
blind tl 
Besides, the 
eyes to the A 
proper ab 
medium was put 
were obt 
folio 
ws : 
However, fairly satisfactory 
5068.9Gt 
5QG8.950 
50CS.957 
7G21.310 
7G21.255 
7G21.2S2 
5083.530 
5083.520 
5083.525 
5090.981 
5090.962 
5090.971 
5105.737 
5105.716 
5105.726 
good . 
Each of these is the mean of many readings, and the agreement is not specially 
