ig20. No. 2. 



ON THE X-RAY SPECTRA. 



Table XIV 



The agreement is seen to be remarkably' good, and thus we see 

 b)' comparison with Paper I. that also in the case of the L-series the 

 assumption of recombination between primaries gives a better agreement 

 for the second line of the series than the hypothesis of recombination 

 from secondaries in the form in which this hypothesis was taken in my 

 previous paper. 



In this way we have succeded in giving a fairly satisfactory explana- 

 tion of the two doublets {re ß) and {y ô). If our interpretation is right 

 they should — in analogy with the K-series — rather be called [uq «1), 



ißoM 



With regard to the /-series we have seen that the assumption of 

 recombination between primaries is not consistent with the explanation 

 previously given to the /-series; but this fact alone must not be considered 

 as fatal to the assumption of recombination between primaries. Further 

 investigations may show us new possibilities for the production of lines, 

 so e. g. it is not excluded, that a ring system — besides the circular 

 and elliptic state — ma)' take up some third state. 



II. 



On the recombination from Secondaries. 



^ II. 



In the Paper I. most of our calculations were carried out on the 

 assumption that the recombining electron did not come from an atomic 

 ring-systems, but from independent stability circles (socondary circles). 



It was further assumed that the recombination towards a broken 

 system took place as if the exterior atomic rings had been removed. 



