230 



CARNEGIE INSTITUTION OF WASHINGTON. 



without some indication of the characteristics of bipolar groups. These 

 are usually exhibited in the form of flocculi following the spot or, in 

 more exceptional cases, preceding it. It seems quite clear that the 

 bipolar spot is to be regarded as the dominant type, of which the uni- 

 polar type may be considered a variant. 



The system of recording described in the above paper has been used 

 daily throughout the year by Messrs. Ellerman, Nicholson, and 

 Brackett. The 307 drawings thus obtained on as many days give 

 the approximate positions, polarities, and field-strengths of all spots 

 large enough to be studied magnetically. 



The accompanying table, prepared by Mr. Nicholson, indicates the 

 polarities of sun-spots observed in the northern and southern hemi- 

 spheres of the sun during the calendar year 1918: 



INCLINATION OF THE LINES OF FORCE IN SUN-SPOTS. 



In the last annual report a method was described for finding the 

 inclination of the lines of force in an east-and-west plane through the 

 center of a sun-spot. Another method, employed by Mr. Nicholson 

 during the past year on 20 different spots, shows that the inclinations 

 of the lines of force in the north-and-south plane through the center of 

 the spot are sensibly the same as in the east-and-west direction at the 

 same distance from the axis of the spot. 



This method is based on the fact that when light from a Zeeman 

 triplet is examined in a direction normal to the lines of force, the vi- 

 brations producing the p-component are parallel to the lines of force. 

 When a sun-spot is observed near the sun's limb, the lines of force 

 lying in a meridional plane through the center of the spot are nearly 

 normal to the line of sight, and their inclination to the north or south 

 can be found by determining the direction in which the light of the 

 7?-component is vibrating. This is accomplished with the 75-foot 

 spectrograph by using a half-wave plate placed over the fixed Nicol. 

 The half-wave plate is mounted on a circle graduated in degrees, so 

 that the angle between its principal axis and that of the Nicol can be 

 observed. In this way the direction of vibration of the p-component 

 projected on a plane perpendicular to the line of sight can be deter- 

 mined in aU parts of the spot. 



