Spectrum oj Erbium. 165 



The aim of X-ray spectroscopy is then to compile a table of the 

 frequencies for all the critical absorption edges of each element rather 

 than tables of emission line wave lengths. Sommerfeld^ proposes to 

 call such a table (values of v/B. when R is Rydberg's frequency 

 .3.29 X 1015 sec-i) a "term" table. This table will contain for the 

 heavier elements 1(k) + 3(l) + 5(m) + 7(n) + 5(o) = 21 "terms," 

 from definite combinations of which we can determine the values of 

 V /R for every emission line for each element. 



Up to the present terms for elements of atomic numbers 61 — > 73 

 have not been measured, and the purpose of this paper is to show how 

 the K "term" of erbium (n = 68) was determined; soon it is hoped 

 to be able to give the K " terms " for the remaining earths. 



X-Ray Spectroscopy. 



The analysis of an heterogeneous X-ray beam into its homogeneous 

 constituents is made possible by the " three dimensional " grating 

 formed by the arrangement of the atoms in a crystal. Calcite was 

 chosen as grating crystal because of its relatively large reflecting power, 

 and because it is easier to obtain a calcite crystal which is a perfect 

 X-ray grating than a rock-salt crystal. The grating space " d " is 

 such that with the spectrometer used, the Ka X^' lines of tungsten 

 (A.Ka'- AKtt = 4.81 X 10—11 cm.) q^yq resolved with a slit width equal to .12 

 mm. 



A full description of an X-ray spectrometer, similar to that used, 

 was given by Mr. Rogers in these Proceedings, May, 1922, so that here, 

 only those modifications will be described which have been added in 

 an attempt to increase the accuracy of the determinations, and facility 

 with which the apparatus can be used. 



The wave length of the erbium K critical absorption edge is ap- 

 proximately .217 X 10 -*^ cm., so that Bragg's fundamental equation 

 for the diffraction of X-rays by a crystal, 



X-2a^me (1) 



can be written, since the angle of reflection of this wave length amounts 

 to approximately 2^, 



\ = C.O.. (2) 



i.e., the wave length is proportional to the angle of reflection. 



When photographing the K absorption edge the crystal was made 

 to oscillate about its axis through 7 mins. of arc by means of a cam 

 driven by an electric motor and reduction gear. This was necessary, 

 as exposures of nearly seven hours were required to obtain a suitable 

 photograph of the edge. 



A crystal holder was designed so that the crystal could first be 

 made vertical and then brought into the axis of rotation. In fig. 1. (a) 

 the crystal C is held rigidly by a piece of rectangular tubing A, sup- 

 ported by the main carrier B, at the three points B^, B^, B^, by loosely 

 fitting bolts, rigidity being obtained by three very stiff springs, 



s , s , s . 



l' 2' 3 



