AURORAE AND MAGNETIC STORMS 
1. Altitude Effects. In all auroral forms there is a 
distinct increase in the intensity of violet nitrogen 
bands relative to the green auroral line as one goes 
Tasue I. AurorAL Lines AND Banps 
(After Vegard [15]) 
A Intensity Identification 
7906 1PG 7-6* 
7867 etl — _ 
6619 == = 
6605 40 1PG 6-3 
6592 il JP 13-11 
6363 = Ox (1D, — §P1) 
6300.30 28 O1 (LD. — 3Po) 
5990.8 15 1PG 15-12 
5891 13 1PG 9-5 (Na: DiD2) 
5577.35 100 O1 GSo — 1De 
5238 6 enG: 16-11 
4709 8 NG 0-2 
4652 5 NG 13 
4596 3 NG 2-4 
4551 2 NG 3-5 
4278 24 NG Q-1 
4236 6 NG 1-2 
4059 3 ® IP & 0-3 
3998 4 2PG 1-4 
3942 2 2PG4 2-5 
3914 47 NG 0-0 
3805 5 2PG 0-2 
3755 4 2PG 1-3 
3578 10 DeEAG: 0-1 
3537 5 2PG 1-2 
3466 = Mt GS = BP) 
3371 9 ® IP 0-0 
3159 6 2PG 1-0 
3126 4 2PG 2-1 
* Vibrational quanta numbers for the transition. 
from the lower edge to the upper limit of the aurorae. 
There is also a distinct increase in the relative in- 
tensity of the red doublet at 6300 A compared to the 
green line 5577 A. At the same time the red nitrogen 
bands 1 P G at 6500 A are relatively stronger at the 
lower part of the border than at the upper limit, when 
compared with the green line. 
2. Type Effects. Forms such as diffuse and pulsat- 
ing surfaces, compared with strong and radiant forms 
such as draperies, bands, and arcs, show violet nitro- 
gen bands of considerably greater intensity than the 
green line. In the sunlit aurorae, where the upper limits 
may attain heights of 800-1000 km, the violet region 
is also strongly enhanced relative to the green line, 
and the color may appear to be gray-violet. 
The red coloring of aurorae is due either to an en- 
hancement of the red oxygen line 6300 A or to the red 
nitrogen bands at 6500 A. The former effect occurs 
especially at lower latitudes; the red coloring at the 
lower edge of strong aurorae is usually due to the en- 
hancement of the nitrogen bands. When the aurorae 
appear in uniform red-colored forms, especially at low 
latitudes, the coloring is due to the oxygen line. How- 
ever, the red color of the lower edge of strong aurorae 
is due to the nitrogen bands. 
301 
3. Latitudinal Effects. These effects were summarized 
by Vegard in a comparison between 70°N (Tromsé) 
and 60°N (Oslo), as follows: (1) The intensity of 6300 A 
relative to the green line increases towards lower lati- 
tudes. This relation is even more pronounced when the 
green line is compared with the negative bands. (2) 
The intensity of the green line relative to the negative 
bands increases towards the lower latitudes. 
Appearance of Hydrogen and Sodium Lines. Among 
the fainter lines in the auroral spectrum are the two 
Balmer lines, Ha (6563 A) and Hf (4816 A), which 
may appear with varying intensities. Furthermore, the 
Na-line (5891 A) also appears. This line is strongly 
enhanced in the sunlit part of the night-sky spec- 
trum. 
Determination of Temperature from the Nitrogen 
Bands. According to the quantum theory, the width 
of the nitrogen bands depends on the temperature of 
the gas. By measuring the intensity distribution within 
the R-branch of a band, the apparent temperature of 
the gas can be calculated. In laboratory experiments 
these measurements are based on spectra taken with 
great dispersion. In auroral spectroscopy, the disper- 
sion is limited, and the accuracy is accordingly limited. 
Quantitative measurements by Vegard based on the 
photometry of the nitrogen band 4278 A give a tem- 
perature of about —45C. 
The Corpuscular Theory of the Aurorae 
The coincidence between the appearance of mag- 
netic storms, aurorae, and sunspot activity leads natu- 
rally to the assumption that the primary cause of 
both magnetic storms and aurorae must be a corpuscu- 
lar radiation emitted by the sun. The fact that aurorae 
appear on the night side of the globe may be explained 
by the effect of the earth’s magnetic field on an elec- 
trically charged stream of corpuscles. Different views 
have been expressed as to the nature of the particles 
emitted by the sun, and at the present time we have 
no definite evidence as to whether they are electrons, 
positive rays, or a mixture of both held together by 
electrostatic action. 
Stérmer [14] has treated extensively the case of the 
movement of an electrically charged particle approach- 
ing the earth’s magnetic dipole field. The theory can 
be applied to both negative and positive charges, but 
it has been especially worked out for the assumption 
that corpuscles are fast electrons. Chapman [3] and 
later Alfvén [1] developed a corpuscular theory in 
which a cloud of negatively and positively charged 
particles leaves the sun at a comparatively moderate 
velocity. This ion cloud is polarized in the earth’s mag- 
netic field, and electrostatic fields are set up. The de- 
tailed analysis of the orbits of the particles is difficult 
because of the complexity of the problem. 
The Model Experiments of Birkeland. A first approach 
to a corpuscular theory of aurorae and magnetic storms 
was made through the model experiments of Birkeland. 
A small uniformly magnetized sphere, the ‘Terrella,”’ 
was placed in a vacuum and a stream of electrons was 
directed against the sphere. At low pressures the paths 
