THE EVER-WIDENING WORLD OF STARS. 47 



noble mirror would require to be increasad a hundred- 

 fold in power before we could see them. For clearly 

 the nebulae, which appear as mere tiny specks upon the 

 vault of heaven, must be very much farther away than 

 the confines of our system, if they are comparable with 

 it in size. 



Therefore we must have ' of two things one.' Either 

 the confines of our sidereal system are constituted very 

 differently from the parts in our neighbourhood; or 

 the nebulae are constituted very differently from the 

 sidereal system. We say, of two things one, meaning 

 that one of the two views must be true ; but it is plain 

 that there is nothing to prevent both being true. 



We may next come to the inquiry whether these 

 views are severally supported by any special evidence. 



Now as to the first, it happens that the southern 

 heavens surveyed by the younger Herschel afford 

 evidence such as Sir William Herschel was not pos- 

 sessed of. The former has seen places in the southern 

 skies where the outline of the Milky Way is so sharply 

 defined that even in the telescope the sudden change 

 from a background of black sky to the sprinkled light 

 of the galaxy is not lost. One half of the field of view 

 will exhibit the former aspect, the other the latter. 

 Now if we consider a cloud, or a dense flight of birds, 

 or any cluster of objects exhibiting a well-defined out- 

 line, we see at once what that well-defined outline 

 means. It signifies that the eye is directed along the 

 'edge or surface of a distinct cluster of objects in one 

 case globules of water, in another birds, and so on 



