Spectrum of Erbium. 



16' 



Adjustments. — The plane of the tube slit S^ was first adjusted by 

 an optical method to be perpendicular to the radius O S^ (Pig. 1.) and 

 at the same time vertical. 



A reflected image of S was observed in the polished crystal face. 

 The crystal was adjusted as explained above so that the slit S^ and 

 its image were parallel for all positions of the crystal. 



To set the surface of the crystal in the axis of rotation a fine 

 needle point, lying horizontally opposite the middle of the crystal, 

 was observed by a long focus microscope. The needle was adjusted 

 until the point did not move in the field of view when the table was 

 turned through 360° The crystal was brought up until the reflection of 

 the point, and the point itself, just touched when seen under the micro- 

 scope. In this manner it was possible to set the crystal rapidly so 

 that its middle section coincided with the axis of rotation to within 

 rb .0005 cm. 



In order to ensure that the central ray from the target passes 

 through the axis of rotation, the crystal was set at zero and an X-ray 

 photograph taken of the crystal. If the setting of the target is true, 

 the edge of the crystal lies exactly in the centre of the darkening, 

 which is limited by S^. 



Experiment. — As the erbium K absorption edge could not be dis- 

 tinguished from the Ka' line of tungsten, it became necessary to use 

 a Gundelach gas tube with a platinum target as a source of general 

 radiation . 



Two millimetres thickness of erbium oxalate were placed before slit 

 S , and the absorption edge registered on one half of the film, while the 

 reference tungsten K a K a' lines were placed on the other half. The 

 distances Wxa ~> Efkl, Wna -^ Wna' were measured by project- 

 ing the film (magnification 10) on to a vertical platform, motions of 

 which could be read by a dividing engine screw accurately to .0005 mm. 

 Table I. contains an actual series of displacement measurements.-^ 



TABLE I. 



""a 



.:! 



M 



"cj 



:<! 



DO 



M 



x 





a: 



s 





& 



u 



1 



X 



^ 



t 



2 



B 



t 



t 



t 



X 



e 



CJ 



^^ 



5J 



tt 



^ 





pa 



^ 



^ 





>mm. 



mm. 





mm. 



mm. 





1-513 



2.353 



216-08 



1-497 



2-346 



21614 



1-523 



2-398 



215-96 



J -488 



2-343 



21617 



1-524 



2-386 



215-99 



1-503 



2-349 



21612 



1.514 



2-356 



21609 



1-508 



2-358 



216 12 



1-515 



2-366 



21609 



1-497 



2-363 



21619 



1-512 



2-350 



21607 









* The method, suggested by Profes-sor Laby, developed in this laboratory for measuring these 

 small displacements has been described previously by J. S. Rogers, M.Sc. (I.e.). 



