162 : 
NPI Oe 
| DecEMBER 13, 1906 
We have no time, however, to discuss this further, 
because I should like to refer to the important subject 
of the 
Behaviour of the Different Lines in the Magnetic Field. 
In many metallic spectra a number of lines occur which 
are closely related, and form so-called series of lines. The 
important discoveries of Hartley, Liveing, and Dewar were 
followed by the discovery of series, owing to the in- 
defatigable efforts of Balmer, Kayser and Runge, Rydberg 
and Schuster. 
The plate shows diagrammatically the arrangement of 
the three connected series which are found in the spectra 
of the alkalies and other elements, 
and which are distinguished by 
Prof. Schuster’ as the trunk series 
(Kayser and Runge’s ‘‘ Haupt- 
serie’), the main branch series 
(Kayser and Runge’s ‘* Zweite 
Nebenserie’’), and the side branch 
series (Kayser and Runge’s “ Erste 
Nebenserie ’’). 
The laws of these 
simpler than those governing 
acoustical vibrations. They are of 
an entirely different character; for 
instance, the members of each 
series approach some definite limit 
of frequency, whereas the number 
of acoustical vibrations may increase 
series are 
indefinitely. 
My first measurements already 
made it evident that lines of 
different series behaved entirely un- 
like each other.” Hence the ratio 
of charge to mass could not be the 
same for all vibrating electrons. 
Runge and Paschen have proved 
in a most beautiful and systematic 
investigation® that all the lines of 
a trunk or of a branch behave in 
the same manner. This result was 
first announced by Thomas Preston,* 
but it is not stated to what degree 
of accuracy and for how many lines 
he investigated the subject. 
All lines of the same series are split up in the same 
manner, e.g. all lines are resolved into triplets or all into 
nonets. Moreover, not only the general type of sub- 
division is the same, but even the amount of separation 
when measured in oscillation frequency. 
The second law discovered by these physicists is this: 
That corresponding series of different elements show the 
same type of resolution, and the amount of separation is 
the same when measured on the frequency scale. 
In the alkalies each line of the trunk series is double, 
and we may speak of a 
twin trunk. The yellow 
sodium lines are a typical 
| example. The type of re- 
Fe OER, 
TLR 
Fic. 9. 
solution of the two lines 
is shown in the diagram 
(Fig. 9). Here we have 
again our old sodium lines 
in the field. The same 
division occurs in all cases 
when twin trunks exist. 
Substances so different in 
chemical behaviour as sodium, copper, silver, and calcium 
(e.g. the well-known lines H and K), split up in the same 
manner; and I think that even Sir William Crookes will 
be surprised to hear that his thallium lines are in the 
magnetic field only counterfeit sodium lines. I can show 
you the splitting up of these beautiful thallium lines in the 
slide. 
1 Schuster, ‘‘ The Theory of Optics,” p. 282, 1904. 
2 “Zeeman, Verslagen Ak. vy. Wetenschappen, Amsterdam, December, 
Dr 
si es | 
1807. Phil. Mag., February, 1898. ; 
3 Runge and Paschen, Berl. Akad. Abhandlungen, Anhang, 1902. Sitz- 
berichte, Berlin, p. 380, p. 720, 1902. Runge, ‘Physik. Zeitschr.,” 
3. Jahrgang, S. 441. Kayser, Spektroscopie, Band 2, Kapitel ix., 1902. 
4 Preston, Dublin Trans. (2) 7, pp. 7-22, 1899. 
NO. 1937, VOL 75| 
With zinc, cadmium, mercury, and calcium, there are 
three main branches associated with each other. The 
amount of separation is the same in each of these branches. 
The type of resolution is shown in the diagram (Fig. 10). 
I can show you further lines of mercury, the triplet, the 
sextet, the nonet. Another example of the same sextet 
is given by a zinc line. The next slide refers to some 
beautiful magnesium lines exhibiting the same three types 
of resolution (Fig. 11). 
We see that in these cases the simple image of an 
oscillating electron does not apply. I regret to say that 
the electronic theory cannot yet give us the explanation 
of the more complicated resolutions; even for the quartet 
we are yet in want of a 
model. 
The laws discovered, 
however, seem to point to 
the conclusion that all the 
lines of a series are 
emitted by one oscillating 
system, that there are, 
therefore, as many series 
in the spectrum of a sub- 
stance as oscillating 
systems in its atom; 
moreover, that the oscil- 
lating mechanism is the ; 
same in different elements. We are reminded here of the 
view advocated by Sir Norman Lockyer that the different 
elements have something in common. The relation between 
these spectral series and resolution in the magnetic field 
is so close that we may expect that the solution of the 
problem of the series will give at the same time the solu- 
tion of the magnetic separation problem. 
That Lorentz’s theory is on the right track even in the 
case of the more complicated magnetic effects appears from 
the polarisation of the nonet shown in the slide. Three 
groups of vibrating lines here correspond to the three lines 
of the triplet. The circular polarisation corresponds also 
Fic. 11. 
to that of the doublet, indicating that it is always the 
negative electron which executes the vibrations. There is 
yet room enough for experimental work in extending these 
investigations in different directions and to other elements. 
Much light on our present subject will be thrown un- 
doubtedly by the activity in adjacent chapters of physics. 
I can only mention in this relation the extremely interest- 
ing experiments by Lenard and Stark on the centres of 
emission of different spectral series, and the important 
theoretical work by Drude’ on the optical properties and 
electronic theory. Maxwell has said, ‘‘an_ intelligent 
student armed with the calculus and the spectroscope can 
1 Drude, Axnalen der Physik, pp. 677, 936. Bd. 14, 1904. 
