WEIGHING THE WORLDS 



"In the first place,'* says Professor Turner, "stars 

 of a suitable spectral type and magnitude for velocity 

 determinations are not found exactly in the ecliptic; 

 second, the Earth's orbit is slightly elliptical; which 

 makes the Earth's velocity a variable quantity; third, 

 it also rotates on an axis and the observer has a com- 

 ponent of velocity relative to the star depending 

 upon the hour angle of the observation; fourth, the 

 earth and moon revolve about a common center of 

 gravity, and planetary perturbations also change 

 slightly the elliptic velocity of the earth; fifth, some 

 of the larger planets, like Jupiter and Saturn, swing 

 the sun out of position sufficiently to affect the star's 

 motion with respect to the sun's center; sixth, the 

 star may be a spectroscopic binary and have a vari- 

 able velocity with respect to the observer (the com- 

 ponent of the sun's motion through space in the 

 direction of a star is assumed to be constant) ; and, 

 seventh, the observations cannot be taken exactly at 

 quadrature with the sun for a number of plates or 

 spectrograms must be taken of each star at successive 

 quadratures." 



That would seem to make the problem almost 

 hopelessly complex. But Professor Turner hastens 

 to assure us that allowance can be made for all these 

 departures from ideal conditions, and that when such 

 allowance has been made, with the resources of the 

 modern mathematician, the resulting value of the 

 sun's distance has a very high degree of accuracy. 

 Reviewing a piece of work started by Sir David Gill 

 at the Royal Observatory, Cape of Good Hope, and 

 completed under Director S. S. Hough, in which the 



79 



