MoNCK — The Distrihution of the Stars in Space. 471 



as unity, tlie number in the successive interspheral spaces will be 

 7, 19, 37, 61, &c., the general expression being {n + ly - w^ 

 Each of these sets of stars will (on the hypothesis of uniformity) 

 give us the same amount of light. 



One obvious consequence follows from the foregoing considera- 

 tions. If the radius of the sphere becomes infinite, the quantity 

 of starlight becomes infinite also. Strictly, of course, this could 

 not be true. One star would get in the way of another, and there 

 are probably a number of dark bodies in space which would in- 

 tercept some of the light without giving any themselves. But 

 after making a very liberal allowance for this kind of obstruction, 

 the entire sky ought to glow with a brightness exceeding that of 

 the full moon. There seem to be but two possible reasons why 

 something approaching to this state of things is not found to exist 

 by observation. The first is, that all the stars belong to one vast 

 system occupying a particular region of space, and that beyond 

 this region there is a practical vacuum The other is, that there 

 is some medium widely diffused through space which intercepts 

 a portion of the starlight, and produces very sensible effects in the 

 case of extremely remote stars. The motions of Encke's comet afford 

 some reason for believing in the existence of such a medium, and 

 other reasons will, I think, appear before the close of the present 

 Paper. 



If we suppose the inner sphere of which I have spoken to in- 

 clude twenty stars, these stars will, on the hypothesis of unifor- 

 mity, be the twenty stars of the first magnitude recognized by 

 astronomers. The second, third, fourth, &o., sets will consist of 

 140, 380, 740, &c., stars in the descending order of brightness, 

 and the light given by each set of stars will be constant and equal 

 to twenty, if we represent the average light of a first magnitude 

 star as unity. Now, Dr. Ball, the Astronomer Royal of Ireland, 

 in his Elements of Astronomy, has given us the number of stars 

 of each magnitude from the first to the ninth, and also several 

 determinations of the ratio of the light of the stars of any magni- 

 tude to that of the magnitude next above it — a ratio which appears 

 to be nearly constant. This ratio has been variously determined, 

 the figures ranging from 0"346 to 0"464. I take 0"4 as the mean, 

 which cannot be far from the truth. The number of stars of each 

 magnitude after the first are, according to Dr. Ball, second mag- 



