Vol. 6, 1920 
PHYSICS: DUANE AND PATTERSON 
511 
from which we compute the wave-length, X, by means of the usual formula 
X = 2J X sin d, 
where d is the distance between the reflecting planes of the crystal. The 
glancing angle for the third and w^eakest absorption band has been esti- 
mated by careful comparison with the neighboring stronger absorption 
drop obtained in the same series of readings. 
Data Obtained. — Table 1 contains the results of our measurements. 
For purposes of comparison we include also the wave-lengths given by 
Wagner^ and de Broglie.^ The angular breadths of the drops vary from 
X-RAY ABSORPTION SPECTRA 
L SERIES 
>5 
0-7 
I 
0.4 
i 
o 0.3 
GLANCING 
MERCURY 
8" /S' 35" 
T 54 40" a. 
MERCURY 
ANGLES 
THALLIUM 
7^39' 15" 
7" 38' 35" 
THALLIUM 
00' OS ICf IS' 20' 25 30' 
6S'> 
CRYSTAL 
55' 00' 05' /O' J5' 20' 25' 30' 35' 40' 45' 50' 
sr 
TABLE ANGLE 
FIG. 2 
three to ten minutes of arc according to circumstances, and we believe 
that we can estimate the positions of their centres to within ten to thirty 
seconds, depending upon the magnitudes of the drops and the regularity 
of the curves. Considerations of this kind have given us the estimates 
of the accuracy of each nleasurement recorded in the table. 
Relative Positions oj Emission and Absorption Wave-lengths. — In the 
article referred to above^ the authors have called attention to the fact 
that in the case of tungsten: (a) the absorption wave-length, a^, is slightly 
shorter than that of the highest frequency emission line of the L series, 
namely, 74; (b) the absorption wave-length, 02, is a large fraction of one 
per cent longer than that of the emission line, 72; and {c) the absorption 
wave-length, Oi, is a large fraction of one per cent longer than the emission 
line, /Ss. The measurements on which these conclusions were based were 
made under the same experimental conditions for both the emission and 
the absorption lines. We have not measured the L series of emission 
