Vol. 8, 1922 
PHYSICS: CLARK AND DUANE 
91 
the best results can be obtained by turning the X-ray tube so that the 
rays that pass through the instrument leave the target almost grazing its 
surface. This gives a narrow beam of rays. We set the ionization cham- 
ber so that the line from the axis of rotation to the chamber's slit makes 
an angle, 26, of convenient size (usually about 10°) with the direct beam 
of rays. Keeping the ionization chamber always in the same position we 
turn the crystal around by successive small steps and measure the ioniza- 
tion current at each step. When a set of parallel planes in the crystal 
lie at an angle 6 to the incident beam of rays the planes pick out of the 
beam X-rays of certain definite wave-lengths, X, and reflect them into 
the chamber, and we get a marked increase in the ionization current. 
Evidently the angles between the positions of the crystal that corre- 
spond to maximum ionization currents are the angles between the crystal 
planes. 
FIGURE 1 
W 3^" W W 
The curve in figure 1 represents the ionization current as a function 
offthe position of the crystal in an experiment with potassium iodide. 
The salt came from Professor Baxter's laboratory, and was of exceptional 
purity. The peaks on the curve are sharply marked, and their angular 
positions can be estimated to within less than 0°.01. The angles between 
several pairs of peaks are 44°.96, 18°.37 and 26°.52, which correspond very 
closely with the angles between some of the principal planes in a cubic 
lattice, namely, 45°, 18°.43 and 26°.57. This together with a similar ex- 
