172 



ANNUAL KEPORT SMITHSONIAN INSTITUTION, 193 8 



behavior of the sun. Of course the spots are fundamentally due to 

 atmospheric disturbances in the surface layers of the sun. Whether 

 the cause of these disturbances is to be found entirely within the sun 

 itself, or whether there are extraneous causes for the phenomenal 

 changes in solar behavior, we do not yet know. The periodic nature 

 of the recurrences of sunspots has suggested to many that planets 

 in some way are the disturbing bodies, but thus far predictions of 

 sunspots based on planetary cycles have not met with much success. 

 Any complete theory of sunspots must obviously express not only 

 why they arise at more or less irregular intervals of about 11 years, 

 but also why the spots of a new series first appear at high solar lati- 

 tudes. It must also account for the progression of the spots toward 

 the solar equator as the cycle advances. Furthermore, since it is 

 definitely known that the magnetic character of the spots changes from 



150 



100 



50 



1910 1915 1920 1925 1930 1955 1940 1945 1950 1955 

 Figure 7.— Clayton's prediction of sunspot maxima based on 68-year period. 



one cycle to the next, an explanation is needed for the change in 

 polarity of sunspots in alternate cycles. Professor Bjerknes, a Nor- 

 wegian meteorologist, has attacked the problem on theoretical grounds. 

 He supposes that there is an atmospheric circulation taking place 

 between the outer surface of the sun and the interior in such a way 

 that in the outer layer of the solar activity, the gases are flowing from 

 the poles to the equator. Near the equator he conjectures that they 

 fall below the surface and then travel northward to high latitudes 

 again. Gaining in temperature through their subsurface migration, 

 these currents rise again to the surface around latitude 40°, proceed 

 once more toward the equator, losing temperatures by radiation, 

 until they fall again into the interior when the solar equator is reached. 

 Accompanying such a circulation on a gigantic scale, the author of 

 this theory postulates that secondary whirls will arise extending into 

 tubular formation. Occasionally forced to the surface, the tube 

 breaks into two segments, the ends of which appear as a pair of sun- 

 spots. A single spot would be the result of but one end of the tube 

 appearing at the surface. 



To account for the change in polarity in alternate cycles, he supposes 

 that while one tubular formation is in progress, another vortex rotating 

 in the opposite direction exists in the lower region and is carried for- 



