156 ANNUAL REPORT SMITHSONIAN INSTITUTION,, 19 42 



distributed with respect to the maximum occurrences of sunspots. 

 The fewest number of aurorae appear to occur from 4 to 6 years 

 before or after the years marking sunspot maxima. The time when 

 aurorae appear most frequently would seem to be about 2 years after 

 the passing of the maximum of sunspots. These results corroborate 

 rather well those of a longer series of observations tabulated by Dr. 

 Chree extending for over 100 years, or from 1750 to 1877. 



The fact that aurorae occur with far greater frequency during 

 years when sunspots are more numerous than during the years when 

 there is a scarcity of sunspots suggests that the electrical effect in 

 the upper atmosphere is something for which a disturbed solar 

 surface is responsible. There is, I believe, a good reason for the 

 fact that the maximum in the auroral displays occurs a year or two 

 after the year of most sunspots. As sunspots begin to wane in 

 numbers, they are nevertheless occurring in regions progressively 

 nearer the solar equator, and as the sun's equator is inclined but 

 slightly to the plane of the earth's orbit, we may draw the inference 

 that sunspots are most effectively associated with the occurrence of 

 aurorae when, other things being equal, they are most nearly in the 

 geometrical plane that the earth travels in its journey around the 

 sun. 



Much of our present knowledge of aurorae is due to the exhaustive 

 studies and mathematical calculations of Dr. Stormer, of Blindern, 

 Norway. By careful analysis of the motion of charged particles 

 in the magnetic field of the earth, he has been able to deduce tracks 

 of ionization so simulating auroral forms as to indicate very signifi- 

 cantly that such discharges in the upper atmosphere are indeed the 

 result of bombardments of electrons coming in from outside, warped 

 by the magnetic field of the earth. In endeavoring to express such 

 phenomena on an electronic hypothesis we may well look at the sun, 

 therefore, for a consideration of the character of sunspots and so 

 trace any possible mechanism by which corpuscular charges might 

 be ejected in the region of the sunspots themselves. 



When we look at an enlarged view of a sunspot and analyze the 

 light from it, we find that the dark interior center is surrounded 

 by a turbulent area. Photographs taken in the light emitted by 

 hydrogen at a particular frequency reveal that there are whirling 

 masses of gas, arranging themselves in veritable vortices. There is 

 every indication, then, that a sunspot is in reality a terriffic solar 

 hurricane. It was in 1908 that the late Dr. George Ellery Hale, 

 the founder and director of the Mount Wilson Observatory, first 

 observed that sunspots were giant cyclones in the sun's atmosphere. 

 They are indeed very similar in their formation to the tropical 

 hurricanes that originate in the West Indies and sweep northward. 



With photographic emulsions made especially sensitive to the red 



