MoNCK — On Star-Distribution. 499 



two indications of greater distance thus concurring. Still, if the 

 motions of this system are independent of those of the great 

 mass of the stars in Cassiopeia's Chair, as they appear to be, this 

 mass of faint stars must he much more remote than the brighter 

 stars which form the system. This result is the more remarkable 

 because the fainter stars are unusually dense in this region. 



Further evidence on the same subject is afforded by clusters of 

 stars. Of course if any cluster was brought nearer to us the com- 

 ponent stars would appear to widen out, and though for some time 

 the density of the cluster might be preserved by the springing up 

 within it of faint stars, which were invisible at the greater distance, 

 if it still continued to approach us, a pretty wide separation of the 

 component stars would ultimately be effected. Now it is, I believe, 

 a fact that in every cluster where the component stars are tolerably 

 bright, as in the case of the Pleiades and the Hyades, a wide sepa- 

 ration is effected by the telescope, while in all very dense clusters 

 the component stars are extremely faint. Our densest clusters in 

 fact are not visible to the naked eye at all, and in inferior tele- 

 scopes they appear only as nebulae. It seems difficult to explain 

 the inverse relation between the density of the clusters and the 

 brightness of the component stars on any other hypothesis than 

 that the average nearness of any group of stars increases with 

 their average brightness. 



I may likewise allude to the general distribution of bright and 

 faint stars over the sky. If brightness did not depend on distance, 

 we might expect to find the same proportion existing everywhere, 

 or at all events to find this proportion as variable in one direction 

 as in another — to find, for instance, that if the density of the faint 

 stars increased more rapidly than that of the bright stars in one 

 rich region, another rich region would be found in which this state 

 of things was reversed. This, however, does not appear to be the 

 case. When we pass from a poorer to a richer region I believe 

 the brighter stars are never found to increase as rapidly as the 

 fainter. Their arrangement seems to indicate that they are moving 

 at a comparatively short distance from us, while the great mass of 

 the faint stars lie far beyond them. There is no instance, I believe, 

 in which a really bright star is known to belong to a cluster or to 

 a sy&tem, the other components of which are much inferior to it in 

 brilliancy. Except in the case of the faint companions of a 



