650 Dr. Marine Siegbahn and Mr. A. B. Leide on 



good specimen of crystal hitherto only calcite has been used 

 as lattice. 



The measurements of the reflexion-angles were carried 

 out in the following way: — The tine slit (with gold edges) 

 was adjusted so that the normal from the rotating axis to the 

 photographic plate passed through it. The crystal was placed 

 on the table with its reflecting plane in the rotation axis. 

 After this adjustment the spectrograph is ready for use. 

 The crystal is then turned so that the normal slit-plate forms 

 an angle with the crystal-face approximately equal to the 

 reflexion-angle for the spectral line sought. This can be 

 done in two ways : in one case the plate-holder may be 

 turned to the right side in order that the incoming pencil 

 shall be reflected ; in the other case a turning of the plate- 

 holder to the left is necessary. 



It will be seen at once that the place of the line on the 

 photographic plate depends solely on the situation of the 

 crystal in relation to the fine slit. When the crystal after 

 exposure of the desired spectral line on the first side is 

 turned exactly 2(£ + 180° and a new exposure made, the two 

 lines will coincide. In general the angle is not exactly 

 2</> + 1$0°, so there will be two lines on the spectrogram. 

 The little correction which must be added to the known 

 angle of rotation is readily determined by measuring the 

 distance between the two line-images and the distance from 

 slit to plate. 



A disadvantage of this spectrograph is that it ranges over 

 only a rather small wave-length region for a definite position 

 of crystal and plate-holder. 



This spectrograph is suitable for wave-lengths between 

 2 A.U. to about 05 A.U., and so bridges over the range from 

 the vacuum-spectrograph to the spectrograph for very short 

 waves described in the second part. But as shown in the 

 later paper, good results may be obtained with as short waves 

 as the K group of tungsten (0*2 A.U.). 



As a first proof the new spectrograph has been used for a 

 measurement of the Ka line of copper. The essential 

 reason for this was a desired control of the measurements 

 with the vacuum-spectrograph with waves of these lengths. 

 It may be remembered that the Ou lines were the shortest 

 of the long waves for which the vacuum-spectrograph had 

 been used. 



The result for this wave-length, Ka x of Cu, obtained in 

 the first part of this paper was 



\=1537-36X.U. 



