86 Hinrichs on Dark Lines in the Spectra of the Elements. 
_Greater degree of accuracy can not be attained before we have 
direct measurements of these wave-lengths like those recently 
made by Hisenlohr. The existing facts give these differences as 
a of d=8=0™"-(00008 between the different groups of 
ines. 
The intervals appear not very regular, but smallest (equal to 
0) near the brilliant group III. Extending our view of physical 
combinations (i. e., such as are marked by actually known lines) 
to the whole of the spectrum, we obtain the following intervals: 
Observed, 2:36 223 2952593 7886::32 1 
Sol 
—_ 
Physical, 2:5: 6 BB 2B Sed ' 
+ oF T 6 +, CO, OSS 
both of which are almost as simple as possible. The latter 
would be formed of equal intervals, 6, if the last interval ex- 
tended beyond 641 mm. K, and if some line should be diseoy- 
ered within the last observed interval, i. e., about midway be- 
tween 2650 and 2840 mm. K. Hence the existing observations 
make it very probable that 
IV. The principal corresponding lines or groups of lines are equi- 
distant in regard to their wave-lengths. 
It will be perceived that these successive laws may be con- 
sidered as generalizations of the same principle. Hence it is of 
the utmost importance to see how far other substances give evi- 
dence of the same laws. We will first consider the other alka- 
line earths. 
The spectrum of strontium, as given by Kirchhoff, is too defi- 
cient yet to be of any consequence. 
The darium spectram is more complete, but we have too few 
lines to discern the physical groups; therefore it seems best to 
rely on the more intense lines as corresponding ones,’ We ob- 
tain as before: 
K. G==0""-00042-'914 Differences. 
15 2502-4 
4c 2 is 1 
6b 2461-2 2-8 min 
= 2031" $-4—=38-4-1 
6c 1989°5 62 : 
16 13714 11-0 4840-44 
le 1287°5 i ee 8 
3b 1274-2 12-1 
2a 1083.0 143 oie 
2a 1031-8 15°4 bis 
16 890-2 $-1—-38—-2 
4b 874'3 18°5 
3a 719°6 19°8 13=- 64-2 
2 7187 
oe him teenie ioe pan ow Bey enn 
