©N THE CONSTRUCTION OF THE HEAVENS. 85 



If now the fmall ftar c be added, the perturbations will tin- obfervations 

 doubtedly affed not only the plane of the orbits of the flars, and '"^""j" 

 but alfo their figures, which will become irregular moveable " at p „ re ^/the 

 ovals. The extent alfo of the ofcillations of the flar c will be poffible revolu* 

 affeaed ; and will fometimes exceed the limits c d, and feme- ^ a^Sfve 

 times fall fliort of them. All thefe varieties may eafily be de- force, directed 

 duced from what has been already faid, when Fig. 7 was con- *° a cent » r » 

 fidered. It is however very evident, that this fyftem alfo 

 mull be permanent j fince not only the centre of gravity o will 

 always be at reft, but a o, whatever may be the perturbations 

 arifing from the (ituation of c, will ftill remain equal to 6 o. 



It mould be remarked, that the vibratory motion of the flar 

 c will differ much from a cometary orbit, even though the lat- 

 ter mould be compreffed into an evanefcent elliplis. For, 

 while the former extends itfelf over the diameter of a globe in 

 which it may be fuppofed to be infcribed, the hypothetical 

 attractive force being fuppofed to be placed in its centre, the 

 cometary orbit will only defcribe a radius of the fame globe, 

 on account of its requiring afolid attractive centre. 



After what has been faid, \t will hardly be necefTary to add, 

 that with the affiflance of any proper one of the combinations 

 pointed out in the four laft figures, the appearance of every 

 treble flar may be completely explained ; efpecially when the 

 different inclinations of the orbits of the flars, to the line of 

 fight, are taken into confederation. 



If we admit of treble flars, we can have no reafon to op? 

 pofe more complicated connexions j and, in order to form an 

 idea how the laws of gravitation may eafily fupport fuch fyU 

 terns, I have joined fome additional delineations. TV very 

 fliort explanation of them will be fufficient. 



Fig. 9 (Plate VI.) reprefents four flars, abc and d, ar- 

 ranged in a line ; a being equal to b t and c equal to d. Then, 

 if a o = b o, and c o = do, the centre of gravity will be in o; 

 and, with a proper adjuflment of projectile forces, the four flars 

 will revolve in two circular orbits round their common centre. 

 By calculating in the manner already pointed out, it will be 

 found, that when, for inflance, ao = 1, co = 3, and c = 4 

 = 1, then the mafs of matter in a = b } will be required to be 

 equal to 1,3492. j 



It is not neceffary that the projectile force of the four flars 

 fliould be fuch as will occafion them to revolve in circles. The 



fy fleuv 



