200 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1933 



radial velocities, which will mainly be dealt with in this discussion, 

 vary from zero up to some 40 or 50 kilometers per second, on the 

 average about 20 kilometers, 12 miles, per second, although a few 

 stars have higher velocities, up to 300 or 400 kilometers per second. 



The motions of the stars appear in general to be quite at random 

 without trace of regularity, like a swarm of mosquitoes, for example, 

 or a crowd of people holidaying in a park. It was, however, early 

 realized that our sun, if in motion, would give an apparent syste- 

 matic trend to the motions of the other stars ; and Herschel, as early 

 as 1783, from the proper motions of 13 stars, showed that the sun 

 was moving toward the constellation Hercules. The method used 

 may be illustrated from the example of people moving at random in 

 a park. It is obvious that a person going through the park will see 

 the people in front, on the whole, seem to approach, those behind 

 appear to recede, and those at the sides to be moving backward. 

 In this way, from the proper motions and radial velocities of several 

 thousand stars, we know with some precision that the sun is ap- 

 proaching the constellation Lyra, not far from Herschel's first at- 

 tempt, with a speed of 20 kilometers, 12 miles, per second. Owing to 

 our moving viewpoint on the solar system, it is evident the stars will 

 all have sjDurious apparent motions, and before we can discuss the 

 real motions of the stars we must first of all correct for the solar 

 motion. Hereafter, when speaking of stellar motions, therefore, we 

 shall be referring to the corrected or real motions of the stars. 



The motions of the stars are so nearly at random that it was not 

 until 1905 that Kapteyn found there was a preference of the proper 

 motions for two opposite directions, " star-streaming " as he called 

 it, the effect being as if there were two systems of stars each with 

 random motions, interpenetrating one another. No satisfactory ex- 

 planation of this phenomenon was forthcoming, though it was gen- 

 erally agreed it must be due to the gravitational attraction of the 

 whole system. A second mysterious systematic effect was discovered 

 by Stromberg at Mount Wilson in 1924. Stromberg showed that 

 the motions of all the stars with velocities greater than 80 kilometers 

 per second, the high-velocity stars, were not at random but were all 

 directed to one hemisphere of the sky. Stromberg was unable to 

 offer any satisfactory explanation for this " asymmetry in stellar 

 velocities " as he called it. It was found, however, that the mean 

 direction was almost exactly at right angles to the direction to the 

 center of the galaxy in Sagittarius. 



These two systematic effects in the motions of the stars and the 

 need of some comprehensive treatment of the dynamics of the stellar 

 system have been met only very recently by Lindblad's theory of the 

 rotation of the galaxy in 1926. Although speculations about the 



