i 4 SCIENCE PROGRESS 



estimating the relative vertical and lateral displacements of our 

 sun relative to the sidereal centre, Walkey concludes that the 

 latter lies on the 230° galactic meridian in about 30 S. galactic 

 latitude, which gives a direction passing very close to Canopus, 

 and that its distance from our sun is about iour hundred light- 

 years, although much weight cannot be attached to this estimate. 

 In support of the contention that Canopus is the sidereal centre, 

 it is further necessary to show that it is situated in the plane of 

 the galaxy, and that its distance, as derived from the distances 

 of the opposite galactic streams, agrees with its measured 

 distance. 



This cannot be done at present, because it is impossible 

 to measure the immense distances concerned with any accuracy. 

 The Milky Way may be at any distance from 1,000 parsecs 

 upwards (one parsec being a distance corresponding to a parallax 

 of one second of arc; it is equal to 3*26 light years), and although 

 it is certain that the parallax of Canopus is less than 001", yet a 

 change in the assigned value by a few thousandths of a second 

 of arc one way or the other will halve or double its distance. 

 But although the theory cannot, at the present time, be either 

 conclusively proved or disproved, yet some support can be 

 obtained for it. From a statistical study of the measured line- 

 of-sight velocities of stars, the velocity of the sun in space can 

 be deduced, and also the direction of its motion, though this 

 latter can be derived more accurately from a study of the proper 

 motions. This velocity being known in magnitude and direction, 

 any observed line-of-sight velocity can be freed from the effects 

 of solar motion and the absolute velocity of the star in the line- 

 of-sight obtained. When this is done for Canopus, it is found 

 that, within the limits of experimental error, its line-of-sight 

 velocity is zero. It must here be remembered that the observed 

 stellar velocities are purely relative : the deduced solar velocity 

 is a velocity relative to the mean of the stars or to the mass 

 centre of the system, and in fact different values are obtained 

 from stars of different types. Our whole sidereal universe may 

 itself have a motion of its own through space, just as our own 

 solar system has amongst the stars of our stellar universe, but 

 such a motion can only be detected by means of observations of 

 bodies lying outside our system. If the view that the spiral 

 nebulae fulfil this condition be supported, then when the radial 

 velocities of an adequate number of them have been measured, 



