MOUNT WILSON SOLAR OBSERVATORY 



237 



observer, while at its outer edge they extend in the opposite direc- 

 tion. Thus with a Nicol and quarter-wave plate the red component 

 should be cut off in one case, the violet component in the other. 



All of these effects are easily seen with the 75-foot spectrograph, 

 used in conjunction with the large solar image of the 150-foot tower 

 telescope. When very near the limb, the hues of force directed toward 

 the observer from the inner edge of the penumbra are absent, pre- 

 sumably because of the depression of the umbra and penumbra below 

 the level of the photosphere. Thus a polarity test for a spot in this 

 position would be misleading, as the elUptical polarization would be 

 due to the oppositely directed lines of force on the outer edge of the 

 penumbra. 



SUN-SPOT POLARITIES. 



The remarkable fact that spots which appear before and after the 

 sun-spot minimum are opposite in magnetic polarity was mentioned 

 in the last annual report. At that time the spots of the new cycle 

 were still occurring in high latitudes, but many have since appeared 

 near the equator. With no more than 1 or 2 per cent of exceptions, 

 spots of the new cycle, irrespective of latitude, are opposite in sign to 



those of the old cycle. Thus the assumption that the polarity depends 

 upon the latitude breaks down; it appears to change with the cycle, 

 in spite of the difficulty in assuming that successive cycles differ 

 so markedly. It is barely possible, however, that another general 

 reversal of polarities will occur at the spot maximum, to which v/e 

 are looking forward with interest. 



The excellent Ha plates taken with the 13-foot spectrohehograph 

 have enabled Mr. Nicholson and the Director to make a statistical 

 study of a long series of hydrogen flocculi. The direction of the whirl 

 is given in the accompanying table, which also shows the polarities 

 before and after the minimum. 



The tabular values are numbers of spots for which the polarities and 

 the structure of the Ha floccuU have been examined. V and R in 

 the table heading indicate the transmission of the violet and red 



