J. C. F. Zillner on a new Spectroscope. 61 
The s eries of measurements which was earried out both on 
it quantitatively with such accuracy as appears desirable for an 
approximate (vorlaiifig) control of theoretical conclusions. 
e numbers cited denote the parts of the ae ee 
and refer to the distance between the two sodium li 
Sodium flame. Sun. 
49°5 49°5 
50°5 als 
a3'0 48°] 
49°5 48°9 
50:6+0°6 49°6+-0°5 
In the following series of observations the reversion spectro- 
vith was furnished, not only with another micrometer-screw 
with a somewhat coarser thread, but also with two other systems 
er whose dispersion in the region of the ee line is 
times greater than that of the systems used for the above 
Aseeceagl he Likewise the old achromatic object- shuees of the 
collimator and the observing telescope were replaced xe un- 
af the i ones, ae which not sieloas nothing was lost in sh 
Sun. 
Screw divisions. Deviation from mean. 
6771 0°8 
69°4 15 
6874 +05 
67°9 0-0 
66°6 a 
6671 —1°8 
68:2 +0°3 
68:0 , 01 
69°6 4+1°7 
Mean 67°9--0°3 
According to this, the interval between the two D lines was 
accurately a with a probable error of 33, of its value. 
But in accordance with facts previously presented, a se of 
the distance se the source of light and the spectrose 
with a velocity of four German. miles per — nes effect 2 a 
corresponding displacement of the lines of the 
tra, to : 
the amount of } 1 of the interval of the D lines, a quantity which = 
