CJ82 PHILOSOPHICAL TRANSACTIONS. [aNNOJ785. 



preservation of the whole, though it should not appear to us in what manner 

 this is effected. I shall however point out several circumstances that manifestly 

 tend to a general preservation : as, in the first place, the indefinite extent of the 

 sidereal heavens, which must produce a balance that will effectually secure all 

 the great parts of the whole from approaching to each other. There remains 

 then only to see how the particular stars belonging to separate clusters will be 

 preserved from rushing on to their centres of attraction. And here I must ob- 

 serve, that though I have before, by way of rendering the case more simple, 

 considered the stars as being originally at rest, I intended not to exclude pro- 

 jectile forces; and the admission of them will prove such a barrier against the 

 seeming destructive power of attraction, as to secure from it all the stars be- 

 longing to a cluster, if not for ever, at least for millions of ages. Besides, we 

 ought perhaps to consider such clusters, and the destruction of now and then a 

 star, in some thousands of ages, as perhaps the very means by which the whole 

 is preserved and renewed. These clusters may be the laboratories of the universe, 

 if I may so express myself, wherein the most salutary remedies for the decay of 

 the whole are prepared. 



Optical appearances. — From this theoretical view of the heavens, which has 

 been taken from a point not less distant in time than in space, we will now 

 retreat to our own retired station, in one of the planets attending a star in its 

 great combination with numberless others ; and in order to investigate what will 

 be the appearances from this contracted situation, let us begin with the naked 

 eye. The stars of the first magnitude being in all probability the nearest, will 

 furnish us with a step to begin our scale ; setting off therefore with the distance 

 of Sirius or Arcturus, for instance, as unity, we will at present suppose, that those 

 of the 2d magnitude are at double, and those of the 3d at treble the distance, 

 and so forth. It is not necessary critically to examine what quantity of light or 

 magnitude of a star entitles it to be estimated of such or such a proportional 

 distance, as the common coarse estimation will answer our present purpose as 

 well ; taking it then for granted, that a star of the 7th magnitude is about ^ 

 times as far as one of the 1st, it follows, that an observer, who is inclosed in 

 a globular cluster of stars, and not far from the centre, will never be able, with 

 the naked eye, to see to the end of it : for since, according to the above estima- 

 tions, he can only extend his view to about 7 times the distance of Sirius, it 

 cannot be expected that his eyes should reach the borders of a cluster which 

 has perhaps not less than 50 stars in depth every where around him. The whole 

 universe therefore to him will be comprized in a set of constellations, richly 

 ornamented with scattered stars of all sizes. Or if the united brightness of a 

 neighbouring cluster of stars should, in a remarkably clear night, reach his 

 sight, it will put on the appearance of a small, faint, whitish, nebulous cloud, 



