Solar Vortices and Magnetic Fields

Abstract

SEVERAL weeks ago I discussed with Prof. Hale the matter of solar vortices and magnetic fields, with reference to his recent discoveries, briefly described in NATURE for August 20 (p. 368). It did not seem to me probable that the effects could be accounted for by unequal diffusivity of the positive and negative ions, or by centrifugal separation, nor does it seem necessary to assume, with Prof. Zeeman, that the magnetic effects are due to electrons participating in the vortical motion. Since that time I have recalled that, so far as definite evidence goes, all luminous vapours giving a line spectrum, and therefore capable of showing the Zeeman effect, are positively charged. A flame coloured with sodium or lithium vapour and placed between two condenser plates is attracted by the negative plate. This fact was used by Lenard to determine the velocity of the positive ions. As shown by Riecke and Stark (Phys. Zeit., v., 537, 1904), if a sodium or lithium salt is placed on the kathode of a long spark the coloured vapour remains in that neighbourhood; if placed on the anode it is at once projected across the entire length of the gap. Hemsalech (Comptes rendus, cxlii., 2, 1906) proved spectroscopically that the metallic vapour in the spark is projected solely from the positive electrode. As shown by Stark (Ann. der Phys., xiv., 529, 1904), that part of the luminous vapour in a mercury arc which gives a band spectrum is unaffected by an electric field, while that part which gives a line spectrum is positively charged. The canal and anode rays are likewise examples of positively charged carriers giving a line spectrum.