88 COSMOS. 



relaies to our solar system, whether we consider the rotation 

 of double stars with unequal velocity round one common cen- 

 ter of gravity, or the apparent or true movements of all the 

 heavenly bodies. If we take up the physical description of 

 the universe from the remotest nebulae, we may be inclined 

 to compare it with the mythical portions of history. The one 

 begins L- the obscurity of antiquity, the other in that of inac- 

 cessible space ; and at the point where reality seems to flee 

 before us, imagination becomes doubly incited to draw from 

 its own fullness, and give definite outline and permanence to 

 the changing forms of objects. 



If we compare the regions of the universe with one of the 

 island- studded seas of our own planet, we may imagine mat- 

 ter to be distributed in groups, either as unresolvable nebulaj 

 of different ages, condensed around one or more nuclei, or as 

 already agglomerated into clusters of stars, or isolated sphe- 

 roidal bodies. The cluster of stars, to which our cosmical isl- 

 and belongs, forms a lens-shaped, flattened stratum, detached 

 on every side, whose major axis is estimated at seven or eight 

 hundred, and its minor one at a hundred and fifty times the 

 distance of Sirius. It would appear, on the supposition that 

 the parallax of Sirius is not greater than that accurately de- 

 termined for the brightest star in the Centaur (0"-9128), that 

 hght traverses one distance of Sirius in three years, while it 

 also follows, from Bessel's earlier excellent Memoir* on the 

 parallax of the remarkable star 61 Cygni (0"-3483), (whose 

 considerable motion might lead to the inference of great prox- 

 imity), that a period of nine years and a quarter is required 

 for the transmission of light Irom this star to our planet. Our 

 starry stratum is a disk of inconsiderable thickness, divided a 



* See Maclear's " Results from 1839 to 1840," in the Trans, of the 

 Astronomical Soc, vol. xii., p. 370, on a Centauri, the probable mean' 

 error being 0"-0640. For 61 Cygni, see Bessel, in Schumacher's Jahr- 

 buch, 1839, s. 47, and Schumacher's Astron. Nachr., bd. xviu., s. 401, 

 402, probable mean error, 0"-0141. With reference to the relative 

 distances of stars of different magnitudes, how those of the third mag- 

 nitude may probably be three times more remote, and the manner in 

 which we represent to ourselves the material arrangenient of the starry 

 strata, I have found the following remarkable passage in Kepler's 

 Epitome Astronomies Copernicance, 1618, t. i., lib. 1, p. 34-39: ''Sol 

 hie noster nil aliud est quam una ex fixis, nobis major et clarior visa, 

 quia, propior quam fix a. Pone terram stare ad latus, una semi-diamctro 

 vim lactece, tunc hcsc via lactea apparebit circulus parvus, vet ellipsis par- 

 va, tota declinans ad latus alterum; eritque simul uno intuitu conspicua, 

 qucB nunc non potest nisi dimidia corispici quovis momcnto. Itaque fix' 

 arum sphcera non tantum orhe stellarum, sed etiam circulo lactis versus 

 nos deorsum est terminata.^^ 



