Corpuscular theory of the aurora borealis

It appears that the German physicist, Goldstein,1 was the first to publish the idea that the Sun sends out into space electrical rays analogous to cathode rays, and that we may thus explain the mysterious connection between variations in solar activity and corresponding fluctuations in the magnetic and electric phenomena, on the Earth. Some years later the Danish meteorologist, Adam Paulsen, from his auroral observations in Greenland was led to the hypothesis2 that the aurora is due to cathode rays; but instead of assuming that the rays came from the Sun, he believed that they originated in the upper strata of the atmosphere. Next, in the year 1896 Kr. Birkeland made his remarkable experiments with cathode rays in a magnetic field. He first found that a magnetic pole had an effect on a beam of parallel cathode rays analogous to that of a lens upon a beam of light, viz., to make the rays converge towards a point. This phenomenon led him to the idea3 that the aurora borealîs was due to a similar effect of the Earth's magnetic field on cathode rays coming from the Sun. In order to test his hypothesis, Birkeland exposed a small spherical electromagnet to a stream of cathode rays, and found a series of facts showing analogies to the shape and nature of the aurora. The aurora‐belts in particular were very beautifully produced. These remarkable experiments, which gave the first really good support to the corpuscular theory of aurora, were described on pages 39–42 of the paper “Expédition Norvégienne, 1899–1900, pour l'étude des aurores boréales.4 Notwithstanding that these remarkable experiments tend to show that the aurora is a direct effect of the precipitation of cathode rays in the upper air, Birkeland regarded the aurora as being caused rather by secondary cathode rays produced by strong electric currents in the upper atmosphere;5 in his later publication, however, he arrived at the conviction that the aurora is a direct effect of the electric rays from without.6