260 REPORTS ON THE STATE OF SCIENCE. 



that the difference of velocity of the different types seems too large to 

 account for in such a way, and it is very likely that the suggestion will 

 at once break down when tested ; still it is at present an alternative to 

 the hypothesis that a star's velocity increases with its age. 



There is a certain analogy between the system of stars that we have 

 been considering and the spiral nebulas, which are very numerous in 

 the sky. The view which is, I suppose, a century old has recently 

 been revived by Sir David Gill and others, that these nebulae are great 

 stellar systems like our own. Although it is highly speculative, the 

 idea may help us to a possible conception of our own system. We have 

 the bun-shaped centre consisting of uniformly distributed stars. Around 

 this and in the same plane are coiled spirals, which correspond to the 

 star clouds of the Milky Way. Matter is clearly flowing in or flowing out 

 along the two arms of the spiral — the nebulae are always double spirals. 

 Thus at two opposite ends of a certain diameter of the bun-shaped mass 

 we have opposite currents of matter, and presumably near the centre 

 there must be two currents in exactly opposite directions — -in other 

 words, two star-streams. The crucial test of such a theory is whether 

 these spiral nebulae are sufficiently remote. I understand that evi- 

 dence has been adduced recently to show that the light of these nebulae 

 suffers enormous selective absorption of the more refrangible rays, 

 which has been taken to indicate an enormous distance, placing them 

 far outside our own system. This will doubtless be investigated more 

 completely. If confirmed, this hypothesis opens up to our imagination 

 a truly magnificent vista of system beyond system — of universe beyond 

 universe — in which the great stellar system of hundreds of millions of 

 stars that has formed the main theme of this paper would be an 

 insignificant unit. 



