212 CARNEGIE INSTITUTION OF WASHINGTON. 



INVISIBLE SUN-SPOTS. 



The minor member (usually the following spot) of a bipolar group is often 

 very unstable, sometimes disappearing and reappearing on successive days. 

 After its final disappearance, both calcium and hydrogen flocculi continue to 

 mark its place. These facts, together with the comparative infrequency of 

 single spots and the tendency of spot groups to break out again and again at 

 the same heliographic position, raise the question whether embryo or decaying 

 spots, invisible to the eye, may not be detected on the sun. 



If the umbra is actually present, but insufficiently dark to be seen by the 

 eye, it might conceivably be rendered visible by some device for increasing 

 the contrast, such as photography with ultra-violet light of very short wave- 

 length. Preliminary experiments with this end in view have not proved 

 successful. Or a thermo-couple, used differentially, might measure the slight 

 difference in radiation. But a third method has the advantage of revealing 

 the vortex which appears to be the essential element of a spot, even when 

 there is no change whatever in the intensity of radiation at the point in ques- 

 tion. For the vortex may produce a magnetic field, even when the cooling 

 caused by expansion within it is too slight to affect appreciably the radiation 

 or absorption of the whirling vapors. 



In order to detect the weak magnetic field of such a vortex, a half-wave 

 plate is mounted above the quarter-wave plate and Nicol prism regularly 

 employed for polarity observations with the 75-foot spectrograph of the 150- 

 foot tower telescope. This is caused to oscillate back and forth across the 

 slit by a small electric motor, so that the half-wave plate is alternately inter- 

 posed and withdrawn. If a large spot is on the slit, the red and violet com- 

 ponents of the iron triplet X6173 will then be cut off alternately. The posi- 

 tion of an invisible spot is betrayed by a slight oscillation of the line to right 

 and left, due to the alternate extinction of its red and violet edges. 



Several invisible spots have been found in this way since systematic search 

 for them was begun last autumn. Some of them have followed the disappear- 

 ance of a spot, others have preceded its visible stage, and in still other cases 

 no visible spot has appeared at the point of observation. The field-strengths 

 measured, from 200 to 500 gausses, are of the same order as those of the small- 

 est visible spots. It is hoped that the method may prove useful in the study 

 of spot formation. 



THE p-COMPONENT IN SUN-SPOT SPECTRA. 



Measurements of the wave-length of the p-component in sun-spot spectra 

 have shown considerable changes in position, the line in every case being 

 shifted away from the n-component present on the photograph. The sug- 

 gestion was made by St. John that this result might be due to the influence 

 of the n-component, the two lines forming a pair similar to those for which 

 he had obtained too wide a separation when measured in the solar spectrum. 

 Two different experiments were made by Nicholson to test this possibility. 



In the first experiment, photographs were made of a single-line of the iron 

 arc spectrum through the same spectrograph (75-foot) that was used for the 

 sun-spot spectra. Three exposures were made. For the first and third a 

 toothed occulting bar was employed, so constructed that the exposed part of 

 the slit for the third exposure corresponded to 'the covered part of the first. 

 The middle exposure gave a simple image of the line which should be as 

 straight as the slit itself. Between the exposures the plate was moved by 



