refractive and dispersive Powers of the Atmosphere, S^c. 205 



evanidum totum ,■ atque Stella utrimque coeuntes, si invicem 

 vicince apparent ; rum modica tempestate arva inundant. Si 

 autem nigrescat, rursus veto eodem colore ambce stellce existunt, 

 pluvias significant. Si vero hie (ovo$) qui est e prcesepis borea 

 modice tenebrescens, languide splendeat, cum austrinus asellus 

 lucidus sit, ventum Austrum expecta. Boream vero e contra 

 tenebrescente lucenteque observare oportet. — Arat. Dios. 167. 

 Many corresponding quotations might be added. 

 Some years ago Mr. Barker published a paper in the Phi- 

 losophical Transactions, tending to prove that many of the 

 stars had changed colour in the lapse of ages; some being 

 described as red, which are now white. I do not, however, 

 attach any importance to the remark, because the ancients 

 used the names for colours with the utmost latitude and variety 

 of significations. Rubere, splendescere, purpurascere, and many 

 others only signified to shine brightly*. 



Nevertheless, certain remarkable changes in some stars, 

 and the alternate disappearance and reappearance of others, 

 while some have been actually lost, seem to warrant an opi- 

 nion that the gradual work of destruction and reproduction, 

 so conspicuous through all mundane nature, is likewise going 

 forward on a grand scale among the ponderous systems of 

 worlds which fill eternal space. This consideration does not 

 bear immediately on the subject under discussion, but it ought 

 to be kept collaterally in view. 



Hitherto we have been discussing the probable causes of 

 the colour of the fixed stars shining by their own light, and 

 have supposed the differences of colour to result from the re- 

 spective composition of the light of each. But in considering 

 the planets which shine only by reflecting the light of the sun, 

 we have other things to take into the account. If the planets 

 have no light of their own, the difference observable in their 

 respective colours must arise from a difference in the dispersive 

 powers of their own respective atmospheres, through which the 

 sun's reflected rays may be separated in their passage. We 

 know, indeed, little or nothing about the composition of pla- 

 netary atmospheres; but analogy would lead us to ascribe the 

 variations of their light rather to properties of their surround- 

 ing atmospheres, than to the colour or other qualities of the 

 substance of the planet itself. It may be observed, that the 

 particular position of the Ring of Saturn does not affect the 

 colour of his prismatic spectrum, and therefore probably it 

 throws back the same sort of light as the body of the planet 

 does. Of what are called the Belts of Jupiter, we know almost 

 nothing; but we may conjecture that they may have some- 

 • See Phil. Mag. vol.xlix. p.49,where I have endeavoured to refute this idea. 



thing 



