252 AXNUAIi REPORT SMITHSONIAN INSTITUTION, 1931 



lems that have cropped up since then. But when I turn to the efforts 

 I then made to fit the theory to the observed properties of the sys- 

 tem, it is like a glimpse of the middle ages. Is it possible that only 

 15 years ago we thought the stellar universe was like that ! With some 

 misgiving I found I must place the sun at least 500 parsecs away 

 from the center of the system; but I did not expect to be believed. 

 Nowadays 7,000 parsecs is the minimum estimate. I was perturbed 

 to find that on the rotation theory I must ascribe to our neighbor- 

 hood an orbital velocity as great as 25 km per sec; in excuse I of- 

 fered a suggestion as to how the differential effects of so rapid a 

 rotation might happen to be concealed. Now we admit an orbital 

 velocity 10 times greater, and claim that its differential effects are 

 not concealed but strikingly manifested. It almost looks as though 

 with a little more courage — a little less reluctance to rend the then 

 accepted fabric — the purely dynamical theory might have forecast 

 the changes that have since been made as the result of observation; 

 I doubt, however, if that would have been justifiable without some 

 additional facts to build on. But I take warning not to be in too 

 great a hurry to stretch the dynamical theory to fit even our present 

 enlightened ideas. I seem to hear the voice of the Halley lecturer 

 of 1945 repeating my remark, " Is it possible that only 15 years ago 

 we thought the stellar universe was like that ! " 



In the past 15 years the accepted dimensions of the galaxy have 

 been enlarged tenfold, and we have to start the comparison of theory 

 with observation anew. There is, however, one result which seems 

 to have been able to survive all vicissitudes, viz., the period of revolu- 

 tion of the stars in their orbits around the center. The period 

 adopted in 1914 was 300 million years, which is close enough to the 

 250 million years deduced from the Oort effect. 



STAR STREAMING AND PERMANENCE 



In 1915 the main stimulus to dynamical investigation came from 

 the phenomenon of star streaming — the tendency of the stars in our 

 neighborhood to move to and fro along one particular line rather 

 than at right angles to it. Ever since this was pointed out by Kap- 

 teyn in 1905, it has been recognized as the most conspicious pecu- 

 liarity of the observed proper motions. It might be a merel}'' local 

 arrangement due to two independent clouds of stars meeting and 

 passing through one another; on the other hand it might have a 

 wider significance as part of the general law of motion of the whole 

 galaxy. A suggestion was put forward by H. H. Turner that the 

 line of star-streaming points to the center of the galaxy; if so, the 

 motion seems to be a very natural one. He pointed out that if we 

 examined in the same way the motions of the comets in the solar 

 system they would show a preferential motion in the radial direction 



