WEIGHING THE WORLDS 



ingly brilliant as it is. But he adds that the data 

 as to the orbits of this pair of stars are still rather 

 uncertain, and that future observations may give 

 corrections that will prove our present estimate of 

 the relative masses to be incorrect. 



In explanation of this element of uncertainty, it 

 should be noted that the periods of orbital revolu- 

 tion of the visual binary systems that have been 

 measured, vary from 24 to 196 years, and that the 

 amount of apparent shift in their positions is so 

 excessively minute, even when amplified by our 

 most powerful telescopes, that it is matter for won- 

 der not that there should be uncertainties in the esti- 

 mate, but that it should be possible to make estimates 

 having any degree of sureness. 



SPECTROSCOPIC BINARIES 



The feat of weighing visual binaries, however, 

 does not carry us quite to the limits of the astron- 

 omers' present-day miracles of world-weighing. The 

 final stage is reached in the case of the so-called 

 spectroscopic binaries. These are pairs of stars so 

 closely linked that even the most powerful telescope 

 reveals the couplet only as a single point of light; 

 yet which are proved by a double shift of their spec- 

 troscopic lines these lines alternately showing 

 single and double to constitute two foci of radia- 

 tion, that is to say two stars. The first discovery 

 of a spectroscopic binary was made at Harvard Ob- 

 servatory by Miss Maury in 1889, through observa- 

 tions of photographs of the well-known bright star 

 Mizar. On examining a series of plates, it was seen 



95 



