Precision-measurements in the X-Ray Spectra* 607 



Another spectrogram of: Pt L a gives in the third order 

 the angle 



PtL« i: 3 =13 O 30' 58", 



while the angle of the same line by rock-salt in the first 

 order is 



^ = 13° 27' 47". 



As a result of these two values we find 



rf = 8-409. 10- 8 , 



or as an average of the lattice-constant of potassium ferro- 

 <3yanide 



d= 8-408. 10- 8 . 

 The value which Moseley used was 



^ = 8-454. 10- 8 . 



This implies that the wave-length tables of Moseley 

 must be corrected with —0*54 per cent, in order that the 

 value may be referred to the assumed lattice-constant of 

 rock-salt. 



D. The wave-lengths of the K group in the region Cu-Cl. 



The above methods for accurate wave-length measurements 

 liave been employed for the longer waves in the K group, 

 where the elements Cu-Cl have been investigated. As the 

 spectrograms show, the reflexion of these long waves is 

 approximately a pure surface effect, where the thickness of 

 the reflecting layer may be neglected. In fact, the breadth 

 of the lines on the plate is the same as that of the slit 

 (about 0*1 mm.). The lines have also approximately sharp 

 edges on both sides. The values in the following table are 

 the means of several spectrograms taken after repeated 

 adjustments, and in some cases with different crystals (rock- 

 salt and calcite). The weaker /3-lines are in general referred 

 to the resp. a-line. 



As will be seen from the next paragraph, the /8-line is not 

 single, and that has possibly had no little influence on the 

 result. 



