CXxiv Journal of the Asiatic Society of Bengal. [N.S., XIV, 
deviations observed. Thus a normal triplet may come froma 
single line or from a very close doublet or triplet and appear 
where a complex magnetic type would be expected. 
A highly interesting paper has recently been published by 
Becker on the complex Zeeman effect. Guided by his investi- 
gations on series spectra, Becker has been led to an explanation 
previously suggested by Preston, Drude and others. 
His fundamental hypothesis is that the complex effect ofa 
line giving more than three components is due to the complexity 
in the structure of the parent line, which is supposed to consist 
of a number of components of nearly equal frequency. Each 
of these gives rise to a Zeeman triplet of normal or nearly 
normal type. It is assumed that the number of components 
in the parent line equals the number of p-components, ¢ 
of the latter being accompanied by two s-components. Such@ 
triplet is called a ‘reduced triplet.’’ The application of the 
hypothesis leads to the investigation of two propositions : (1) 
in any complex Zeeman effect the number of s-componen aC 
0 
tw 
others satisfy the first ; another requires further expermme - : 
The Stark Effect. 
was in the autumn of 1913 that Stark obse es x 
It 
effect of an electric field on the resolution of spectrum 
nd 
