ON THE STRUCTURE OF THE UNIVERSE 15 



we may be able to find whether our system as a whole has any 

 motion. 



Remembering this, it follows that the sidereal centre must 

 be at rest relative to the mean of the stars, and the fact that 

 Canopus has no motion in the radial direction, whilst not proving 

 that it is the sidereal centre, is a link in the chain of evidence. 

 If now we assume that it actually does occupy this position, it 

 follows that it can have no velocity in a direction perpendicular 

 to the line of sight, and therefore its observed annual proper 

 motion of '0184" must be parallactic in origin, being an apparent 

 motion due in reality to the motion of our own sun. The known 

 speed of this motion provides a means of calculating the parallax 

 of Canopus, upon the above assumption, and the resulting value 

 is -0067", corresponding to a distance of 149 parsecs. This is the 

 order of magnitude of the quantity to which Gill's direct deter- 

 mination led. This distance, being assumed correct, provides 

 a base line from which to measure other distances in our system, 

 and the resulting distance of the galaxy is found to be about 

 2,000 parsecs, or about 6,500 light-years. We can only say, in 

 reference to this value, that it is of the order of magnitude usually 

 assigned. There for the time being we must leave this theory. 

 The evidence, so far as it goes, is concordant, but it must be 

 left to the future, when, perhaps, some new and indirect means 

 of determining accurately these immense distances may be 

 evolved, to establish it firmly or to disprove it. ► The idea that 

 the centre of our system is occupied by an immense sun, many 

 thousands of times larger and more glorious than our own sun, 

 and that round about it are millions of lesser suns of various 

 sizes, together forming the nucleus of an immense spiral nebula, 

 of which the spiral arms coiling around the nucleus appear to us 

 as the Milky Way, and that this to us immense system is but 

 one, and perhaps a comparatively small, island universe amongst 

 thousands or millions of other island universes in space, is an 

 idea which by its magnificence appeals to the mind of man. 

 Upon what basis of truth this conception is founded I have 

 endeavoured as briefly as possible to elucidate in the present 

 article. So brief a discussion of so great a theme is necessarily, 

 both from limitations of space and from present incompleteness 

 of knowledge, very inadequate, and many important and related 

 phenomena, such, for example, as that of star-streaming, or 

 Russell's theory of "giant" and "dwarf" stars, which changes 



