STRUCTURE OF THE GALAXY — PLASKETT 201 



rotation of the stellar system about some central star were prevalent 

 during the nineteenth century, they had no observational data to 

 support them and could not survive. Before giving Lindblad's 

 hypothesis, however, there are some general considerations unknown 

 earlier in support of a rotation of the galaxy which may well be 

 cited. 



We have already built up our conception of the galaxy as a tre- 

 mendous discoidal aggregation of stars and star clouds with a diam- 

 eter of 200,000 and a thickness of 10,000 light-years. Elementary 

 dynamics at once indicates that such a thin disk-shaped system could 

 only be maintained in a flattened form by rapid rotation, otherwise 

 it would assume a more nearly spherical shape. And we have not far 

 to go to find analogues showing rotation. The very structure of the 

 spirals suggests rotation; they are just like a Catherine wheel. 

 Moreover, the spectroscope has demonstrated that several of them 

 are in raj)id rotation. Unfortunately, however, we cannot get out- 

 side our galaxy to test its rotation spectroscopically, and there did 

 not seem any means of attacking the problem internally from our 

 position within the system, where all the stars we can observe are 

 rotating as well as ourselves. 



The way was first pointed out by Lindblad in 1926 in a series 

 of papers to the Swedish Academy. Lindblad postulated that the 

 galaxy was composed of a number of subsystems, approximately 

 concentric with one another, and all in rotation at different rates 

 around a common axis f>erpendicular to the galactic plane. The 

 subsystem rotating at the highest speed would evidently by centrif- 

 ugal force be the most flattened toward the central plane and, on 

 Lindblad's hypothesis, is the one containing our sun, the local cluster, 

 the Milky Way clouds, and the vast majority of the stars. Sub- 

 systems with a lower rotational speed would be less flattened to the 

 galactic plane. The stars in such a slower-moving subsystem would 

 appear to us to be moving much faster, to have higher radial veloc- 

 ities than the stars in our own system, as our high rotational speed, 

 about 300 kilometers, or 200 miles, per second, would sweep us 

 rapidly past them. The most slowly rotating subsystem is the 

 system of globular clusters which are little flattened toward the 

 plane and whose low rotational speed gives them apparently high 

 velocities in the line of sight with respect to the sun, 



Lindblad's hypothesis at once gives an explanation of Stromberg's 

 asymmetry, as the high-velocity stars evidently belong to a slower- 

 moving subsystem and being overtaken by the higher rotational 

 speed of the sun, at right angles to the direction to the center, gives 

 them the appearance of moving in the opposite direction, as has 

 already been shown. The phenomenon of star-streaming, the pref- 



