1316 
It would be interesting to know, how these lines behave in a 
magnetic field, if they have been excited by other means e.g. by 
those of which Bmnam could dispose. Evidently for this group the 
comparison of ZpEMAN-effect and pressure-effect can be only very 
incomplete. 
Still I thought it suitable to point to the remarkable absence of 
the lines of this group in Miss GRAFTDIJK’s observations. Possibly 
this remark may still be useful for an explanation of the connection 
which in all probability exists between the two effects. 
The incompleteness of the groups IV and V must principally be 
ascribed to the want of quantitative observations of the pressure- 
effect. For these groups the measurements of the Zurmman-effect are 
rather complete. 
Though Brrmam has qualitatively investigated the given lines, which 
has rendered possible the division into groups, he has unfortunately 
not made quantitative measurements on them. Also in DurrieLp’s 
paper we miss most lines of these groups. 
I have still mentioned all these lines, that in any case I might 
be able to say something about the value of the Zneman-effect for 
each of these groups. 
Let us now firstly consider the magnitude of the two effects 
separately and let us begin with the Zevman-effect. 
Then we can directly remark that the different lines which belong 
to one and the same group have not all the same Zueman-effect. From 
the numbers it is also evident that this will neither be the case with - 
da Beets 
the quantity : Pere from which it will be known that in spectra 
showing line-series it is a constant for the lines of the same series. 
Because of the different values of the Zrnman-effect it seems therefore 
for the moment impossible to find spectral series in these groups, the 
existence of which is suggested by the analogous behaviour of the lines. 
Still we can make some remarks on the different groups. Exelu- 
ding group III because there is too-little observation-material, we 
can make up for each of the groups a mean value of the ZremMAn- 
effect. From these values: 380, 405, 529, 678 we see directly that 
the first two groups give the lowest value of this effect. For the 
groups IL and I these values differ only slightly from each other, 
which behaviour we shall find back for other quantities which we 
have calculated for these two groups. From the five groups we 
have distinguished, , these two are the least different ones. With a 
single exception the lines of the groups IV and V show larg« 
ZreMAN-effects. 
