Marcu II, 1915] 

LETTERS TO THE EDITOR. 
{The Editor does not hold himself responsible for 
opinions expressed by his correspondents. Neither 
can he undertake to return, or to correspond with 
the writers of, rejected manuscripts intended for 
this or any other part of Nature. No notice is 
taken of anonymous communications. | 
Resonance of Sodium Vanour in a Magnetic Field. 
Irv is well known from the observations of Wied- 
mann and Schmidt, Wood, Dunoyer, and others that 
dilute sodium vapour contained in a glass bulb emits 
resonance radiation when a soda flame is focussed 
upon it. During the last few days I have examined 
the effect on the resonance radiation of putting the 
resonating vapour in a magnetic field. 
If the exciting flame contains very little soda, the 
resonance radiation is diminished by the field. 
If, on the other hand, the flame is rich in soda, the 
field greatly increases the resonance radiation. 
It is too early to put forward more than a tentative 
suggestion towards the explanation of these effects. 
Each sodium line emitted by the resonating vapour 
is broadened by the Zeeman effect. The flame poor 
in salt gives a narrow exciting line, and magnetic 
broadening throws a part of the resonating line off 
the exciting line, thus diminishing the light. 
Adding more salt to the flame makes each exciting 
line broader, and (it is provisionally assumed) reverses 
the middle of it. Thus magnetic broadening of the 
resonance line tends to bring the brightest parts of 
the exciting line into action, and increases the light. 
It may be remarked that with an intermediate con- 
dition of the flame a moderate field would produce the 
kind of effect last referred to, while a very strong 
field would separate the side components so far as to 
throw them beyond the limits of the exciting line. An 
effect of this kind has been observed, though unfor- 
tunately the condition of the exciting flame at the 
time was not noted. The current was switched on, 
and as the field increased (this takes a perceptible 
time) the resonance radiation increased and_ then 
diminished again. On turning the current off, the 
light again passed through a maximum. The greatest 
strength of field used in this experiment was about 
14,000 units. Brightening can be distinctly observed 
with tooo units, when a well-salted flame is used. 
I have not been able to find that any previous 
observations have been made on the resonance of 
sodium vapour in a magnetic field. Observations 
were made on mercury vapour by Malinowski (Phys. 
Zeits., September, 1913). The present experiments 
were suggested by some made in this laboratory by 
Mr. F. S. Philipps on mercury vapour (see Nature, 
December 4, 1913). His observations were independent 
of Malinowski’s. RAMS URULM. 
Imperial College, South Kensington, March 9. 
The Spectra of Hydrogen and Helium. 
Dr. Bonr’s letter in Nature of March 4, although 
giving an interesting discussion of some aspects of 
this problem, does not meet the particular point which 
my letter was designed to raise. This point was 
solely that since combination series must be expected 
from the ‘‘4686” series in any circumstances, and 
since the lines so calculated occupy the positions in 
which lines have been found by Evans, they cannot 
be used to discriminate between theories of the origin 
of spectra, for we cannot prove that the observed lines 
are not these combination lines. It is true that 
Bohr’s theory involves the combination principle, but 
so also does that of Ritz, who originated the principle. 
My letter (NATURE, February 11, p. 642) took up 
NO. 2367, VOL. 95] 

NATURE 33 


this purely negative attitude, and was not intended as 
a criticism of the theory. It did not even advocate 
a hydrogen origin for the lines. Fowler’s view, that 
the * 4686” series is a 4N series analogous to that 
in magnesium, was, in fact, stated to have more 
evidence in its favour. Whether the origin be really 
hydrogen or helium is not actually relevant to my 
argument. Even if the origin is really helium, it was 
pointed out by Fowler in his Bakerian lecture that 
his results do not formally imply Bohr’s theory. 
Since that time, the writer has published a proof that 
the theory cannot explain 4N series in general, for 
such elements as magnesium. It can only deal with 
helium, and the formal analogy between helium and 
magnesium would weaken, rather than strengthen, 
the theory. 
The greater part of Dr. Bohr’s letter does not bear 
on my original point, for he is seeking to discriminate 
between a hydrogen and helium origin, and between 
his view and kKydberg’s, not directly by Evans’s 
experiments, but by other considerations. At the risk 
of going further from the point at issue, I feel that 
some remarks on these considerations are necessary. 
The references to Rau’s experiments on voltages 
necessary to produce series are interesting, and if 
they have been interpreted correctly—there is some 
doubt of this—they show that the chemical origins 
of the series are those stated by Dr. Bohr, and by 
Stark and others. They show also that the electrons 
in Bohr’s model atoms have the proper angular 
momenta. ‘There are other reasons for believing that 
the relation of the atom to Planck’s h is contained 
in the angular momenta, and such atoms were treated 
by the writer some years ago, but with a different 
kind of emission. Nevertheless, Rau’s experiments 
have nothing to do with the mechanism of spectral 
production, and cannot support any theory of the 
mechanism of radiation. For the. radiation problem 
is quite superposed on any specification of the steady 
configurations of non-radiating atoms. 
The remarks concerning Rydberg’s view proceed 
throughout on the supposition that the usual constant 
p—Rydberg’s phase—is zero in these series. No 
such case is known elsewhere in the whole range of 
spectra. It is quite easy to fit the ‘‘4686"’ series into 
a formula exhibiting it as a principal series of hydro- 
gen, if this constant » is not arbitrarily chosen as 
zero. There are other arrangements of the disputed 
series as hydrogen series which are formally possible, 
but their description would occupy too much space 
here. A full account of the whole problem will be 
published shortly, so that I propose to discontinue 
the present discussion with this letter. Meanwhile a 
protest must be urged against Dr. Bohr’s conviction 
that the spectrum of atomic hydrogen consists solely 
of the Balmer, Ritz, and Schumann series. For MM. 
Fabry and Buisson have shown that a very large 
number of lines in the ‘secondary’ spectrum are due 
to atoms of hydrogen. A correct model of the 
hydrogen atom must account for more emission spec- 
tra than have yet been deduced by Bohr’s theory. 
Finally, I must again state explicitly that my present 
purpose is not to call the theory into question. My 
only concern is to show that no decisive factor has 
yet entered, and that judgment between theories must 
at least be suspended for the present. The importance 
to physics in general of the whole question of spectral 
emission is so great that a hasty decision must not be 
made. And the fact remains that all the present 
experimental results are explicable in widely different 
ways. The test mentioned at the end of my previous 
letter still appears to be an obvious crucial one. 
J. W. NicHOLson. 
University of London, King’s College, March 5. 

‘ 
