422 ASTROMETRY 



But, whatever may be the cause of the phenomenon, whether it 

 be cosmical or even only instrumental, we must expect that the 

 spectroscopic observations of the bright stars will show the phenom- 

 enon of the star-streams less strongly than will the observations of 

 fainter stars; and as our list is made up in great part of such bright 

 objects we must expect to find their influence shown in a somewhat 

 less marked degree than we should be led to imagine from the con- 

 siderations of this lecture, which are based on the whole of the stars 

 down to the ninth magnitude. 



Now this is just what we find to be the case. 



I find, arranging in order of the distances from the nearest of the 

 vertices: 



Real velocity in the line of sight 

 (Kilom. pro sec.) 



The phenomenon is clearly shown. The observed numbers, how- 

 ever, are only nearly 83 per cent and the observed amplitude but 

 72 per cent of the theoretical value. 



A small but independent contribution is furnished by nine stars, 

 of which the radial velocity has been published on account of its 

 unusually large amount. If our theory is correct, the largest radial 

 motions must be far more numerous near the vertices than at a 

 greater distance. 



The nine stars in question are more thickly crowded (if such a 

 word may be used of so small a number) within 43 degrees from the 

 vertices than in the rest of the sky, in the proportion of three to 

 one. 



Taking the evidence for what it is worth, we may say that it con- 

 firms the theory. The proof is not convincing, however, and I wish 

 to express the hope that those who are in the position to test the 

 whole theory by more extensive and more reliable materials will not 

 neglect to do so. 



A few hundreds of stars not pertaining to the Orion stars and fainter 

 ihan magnitude 3.5 must probably be sufficient for the purpose. Even 

 if the test is fairly stood, we have certainly to see in the present 

 theory no more than a second approximation to the truth. We shall 

 have to develop its utmost consequences. 



