ON THE CONSTRUCTION OP THE HEAVENS* $\ 



foregoing calculation of the required (ize of the ftar, which 

 tnuft contain nearly double the mafs allotted to other ftars of 

 the 7th magnitude; of which, therefore; none but this one 

 can be proper for making up the required double ftar. If the 

 ftars of the 8th and 9th magnitudes, of which there will be 896 

 and 1 134, mould be taken in, by way of increafing the chance 

 in favour of the fuppofed compofition of our double ftar, the 

 advantage intended to be obtained by the addition" of numbers, 

 will be completely counteracted by the requifite uncommon, 

 bulk of the ftar which is to ferve the purpofe ; for, one of 

 the 8th magnitude, ought to be mOre than 2 J times bigger 

 than the reft : and, if the compofition were made by a ftar of 

 the 9th magnitude, no lefs than four times the bulk of the 

 other ftar which is to enter the compofition of the double ftar 

 would anfwer the purpofe Of its required brightness. Hence 

 therefore it is evident, that cafual fituations Will not account 

 for the multiplied phenomena of double ftars, and that confe- 

 quentiy their exiftence muft be owing to the influence of forte 

 general law of nature; now, as the mutual gravitation of bo- 

 dies towards each other is quite fufficient to account for the 

 union of two ftars, we are authorifed to afcribe fuch combina- 

 tions to that principle; 



It will not be necefiary to infill any further on arguments Aftualobferva- 

 drawn from calculation, as I (hall foon communicate a feries of tl0n ,n( | ,cate f a 



revolution of 



obfervations made on double ftars; whereby it will be feeh> double ftars 

 that many oftftem have actually changed their fituatio'n with regard round their cen- 

 to each other, in a progrejfive courfe, denoting a periodical revo- 

 lution round eaeh other ; arid that the motion of jbme of them is 

 direct, while that of others is retrograde. Should thefe obfer- 

 vations be found fufficiently conclufive, we may already have 

 their periodical times near enough to calculate, within a cer- 

 tain degree of approximation, the parallax and mutual diftance 

 of the ftars which compofe thefe fyftems, by meafuring their 

 orbits, which fubtend a vifible angle. 



Before we leave the fubjec"l of binary fyftems, I mould re- The fun is not 

 mark, that it evidently appears, that our fun does not enter thus combined j 



i- • i , n r » „ though our fyf- 



mto a combination with any other ftar, to as to form one or t em is not abfo- 

 thefe fyftems with it. This could not take place without our lutelv at reft « 

 immediately perceiving it ; and, though we may have good 

 reafon to believe that oiir fyftem is not perfectly at reft, yet the 

 caufes of its proper motion are more probably to be afcribed to 

 Vol. V. — Junk, G fome 





