SOLAR VORTICES AND MAGNETISM IN SUN SPOTS ABBOT, 331 



from a photograph (T 190) made with the Nicol set at 61° E., shows only the 

 red component of the doublet. No. 3 illustrates the effect of turning the Nicol 

 90° ; only the violet component remains. Other spot lines in these photographs 

 change in a similar way. 



Photographs like these seemed to leave no doubt that the components of the 

 spot doublets are circularly polarized in opposite directions. Since the only 

 known means of transforming a single line into such a doublet is a strong mag- 

 netic field, it appeared probable that a sun spot contains such a field, and that 

 the widening and doubling of tlie lines in the spot spectrum result from this 

 cause. But much remained to be done before the proof could be regarded as 

 complete. 



Since this preliminary work I have made over 200 photographs of spot 

 spectra with polarizing apparatus before the slit. In addition to this col- 

 lection of plates, numerous photographs of spot spectra, some taken with 

 polarizing apparatus by Doctor St. John, and others made without Nicol or 

 rhomb by Mr. Adams and myself, are available for study. These have been 

 used for the investigation described in the following pages. 



REVERSED POLARITIES OF RIGHT AND LEFT HANDED VORTICES. 



If a Nicol is set so as to cut off the violet component of a doublet observed 

 along the lines of force of a magnetic field, reversal of the current will cause 

 the red component to disappear and the violet component to become visible. 

 Reversal of the direction of the current in a magnet corresponds to reversal of 

 the direction of revolution in a solar vortex. If it could be shown, by an 

 independent method, that in two sun-spot vortices the charged particles are 

 revolving in opposite directions, the red components of the doublets should 

 appear in the spectrum of one spot, and the violet components in that of the 

 other, the position of the rhomb and Nicol remaining unchanged. 



Fortunately the spectroheliograph plates indicate the direction of revolution 

 in the solar vortices. The vortices are constantly changing in appearance, 

 and the stream lines are not always clearly defined. Plate 5 is reproduced 

 from a photograph of the sun made by Mr. Ellermau with the 5-foot spectro- 

 heliograph on September 9 and 10. It shows two si)ots, one in the northern, 

 the other in the southern hemisphere, with vortices indicating revolution in 

 opposite directions, if we may judge from the curvature of the stream lines." 

 Portions of the spectra of these spots, photographed by myself on September 

 9, are reproduced in plate 4. No. 1 shows the spectrum of the southern spot, 

 in which the direction of revolution was clockwise, taken with the Nicol set 

 at 29° W. Only the red components of the doublets appear. The northern spot, 

 in which the revolution was counter clockwise, was then photographed (2). 

 Although the Nicol and rhomb remained in the same position as before, the red 

 components of the doublets are now cut off, while the violet ones are visible. 

 During this exposure the slit was kei)t on the western umbra of the northern" 

 spot, which was divided into two parts by a bridge (not shown in the repro- 

 ductions). Another exposure, with Nicol and rhomb as befoi-e, was then made 

 on the eastern umbra of the same spot (3), with results similar to those ob- 

 tained for the western umbra. For the final exposure (4) the slit was kept on 

 the eastern umbra of the northern spot, and the Nicol rotated 90°. As was 

 to be expected, the i"ed components were brought into view, and the violet 

 components extinguished. This spectrum is therefore precisely similar to that 

 of the southern spot, which was taken with the Nicol in the reverse position. 



" Right and left handed vortices have also been found in the same hemi- 

 sphere. 



