206 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1920. 



individual faces of a crystal ; (2) by reflecting them from a crystal- 

 line powder so that all of the possible crystal planes have an equal 

 chance to reflect; (3) by passing X rays through a thin section of a 

 crystal. We shall now discuss very briefly the way in which these 

 three kinds of experiments are carried out and point out the kinds of 

 data that can be obtained from each. 



The first of these diffraction methods may be called the spectrometer 

 method. X rays, preferably of a single wave length, are passed 

 through two narrow slits of an X-ray spectrometer in order to render 

 them parallel ; this beam, after reflection from the face of a crystal 

 mounted upon the spectrometer table, passes into an ionization cham- 

 ber, where its intensity is measured by the magnitude of the ionization 

 which it produces in the gas of this chamber. Such an X-ray spec- 

 trometer is shown in figure 12. The X rays passing from an X-ray 

 tube through the slits S and S ti strike the face of the crystal mounted 

 at C. When the ionization chamber D and the crystal C stand at the 

 proper angles, the reflected beam of X rays enters the chamber D 

 through a mica window by way of the slit S 2 . This chamber is filled 

 with some gas, usually sulphur dioxide or, better, methyl bromide, 

 which ionizes strongly under the action of X rays. 



The case of the ionization chamber is charged to a potential of 

 about 200 volts, while the inside insulated electrode, which is con- 

 nected only with the gold leaf of a sensitive electroscope E, is, at the 

 moment of beginning the experiment, at the potential of the ground. 

 As X rays enter the chamber and ionize the gas within it, the poten- 

 tial of the gold leaf will change. The rate of this change is measured 

 by the drift of the gold leaf and taken as an indication of the in- 

 tensity of the reflected X rays. By properly adjusting the positions 

 of the crystal and of the ionization chamber the relative intensities 

 of reflection from a single face can be obtained for the various orders. 

 This information is usually sought for several faces of a crystal. 



In place of an ionization chamber and electroscope for measuring 

 the diffracted X rays a photographic plate can be used. If the X rays 

 are of a single wave length an image of the slit will appear at the 

 position conditioned by equation (1). If the X rays in the original 

 beam are of more than one wave length, as is usually the case, then 

 for a perfectly constructed crystal — meaning thereby one in which 

 the atoms are arranged with perfect regularity throughout the entire 

 mass — and for perfectly parallel X rays, only the image correspond- 

 ing to one wave-length will appear upon the photographic plate for 

 a particular orientation of the crystal. If the beam of X rays is 

 slightly divergent after passing through the slit of the spectrometer, 

 it can be shown from geometrical considerations that the X rays of 

 somewhat different wave lengths which can thus be reflected will 

 focus upon a surface which is as far from the crystal as the crystal is 



