Januaey 1, 1904.] 



SCIENCE. 



31 



pheric phenomena in its scope. The details 

 of the work of the Weather Bureau in this 

 research are being published as rapidly as pos- 

 sible, but as some time must elapse before 

 this will be completed, it may be of interest to 



Pi^.l. 



make a comprehensive statement of the con- 

 clusions that have been reached. 



The Circulation of the Sun. (Fig. 1.) — The 

 thennodynamic conditions in the sun suggest 



a viscous nucleus extending about half its 

 radius from the center, which is surrounded by 

 a gaseous envelope, the sun's proper atmos- 

 phere. The nucleus is apparently not spher- 

 oidal, but dumbbell in shape, according to the 

 Jacobian ellipsoid of equilibrium, so that the 

 sun is an incipient binary star with two cen- 

 ters of action instead of only one. This result 

 rests upon the following facts : (1) The promi- 

 nence frequency numbers on the surface have 

 two distinct maxima, which move in opposite 

 directions from the middle latitudes, one from 

 latitude 25° towards the equator, as do the 

 sunspots and faculse, and the other from lat- 

 itude 50° towards the poles, in the course of 

 an eleven-year cycle. The cycle begins at 

 minimum with a strong outpouring in middle 

 latitudes, which separates into the two 

 branches mentioned. It is probable that the 

 congested energy of the interior first seeks 

 to escape from the region where the viscous 

 nucleus ends, and that one wave spreads 

 through the gaseous region towards the end 

 of the equator on the surface, while a second 

 wave passes through the nucleus towards the 

 center of the sun. The course of the max- 

 ima points, as shown in the Monthly Weather 

 Review for January, 1903, favors this expla- 

 nation. (2) The distribution of the prom- 

 inences in longitude gives two maxima, lo- 

 cated on two opposite meridians of the sun, 

 as if they sprang from two foci. (3) This 

 division of solar activity is also found record- 

 ed in the distribution of several other products 

 of solar energy in the period of the solar rota- 

 tion, which is 26.68 days on the equatorial 

 plane, as in that of the sunspots and the 

 faculse, in one system of deflecting forces of 

 the terrestrial magnetic field, in the baro- 

 metric pressures and in the temperatures. 

 That would be a good reason, if it exists, 

 why the sun in its rotation should effect dif- 

 ferential impulses throughout the cosmical 

 system. 



The periods of rotation of the sun have 

 been determined in the several zones by a dis- 

 cussion of the prominence numbers, and there 

 is retardation from the equator to the poles. 

 This conforms to von Helmholtz's Case II., 

 derived from the general equations of motion. 



