1806.] 



Proceedings of Learned Societies. 



255 



would double that quantity, it (lilt would 

 fall llioit ol" the change obl'er\ed in ii lln- 

 gleycar. 



Dr. Herfchel infers that the projectile 

 jnotion luult be combined with attraction, 

 and tlie motions of the liars, when regu- 

 lated in this manner, are not unlike the 

 dii'polition by which bodies of the folar 

 fyltem are governed. In coniidcring the 

 probable exillence of a centi'C of attrac- 

 tion, it is obfcrved, that there are two 

 ■«\aysin which a centre of attniction, lb 

 jiowerful as the prel'ent occalion would 

 require, may be conftruCted : the moll 

 liinple of them would be a lin^le body of 

 great magnitude, — this may exill, though 

 we (hould not be able to perceive it by 

 liny fuperiority of lulh-e, as the dccreafe 

 of its light ariilug from its 'ireat dillancc 

 vTOuld hiirdly be coin[)enfated by the fize 

 of its diameter ; but as this is hypothec 

 ticul, it cannot be admitted uito the dif- 

 culhon. 



The fecond way of tlie conftriiction of 

 a very powerful centre, may be the joint 

 attraction of a great numljer of ftars 

 united into one condenfed group : the 

 exiltence of fuch groups has been proved, 

 for, fays Dr. Herfchel, " the nebuki dil- 

 covercd by Dr. Halley in the year 1714, 

 ill which he and other obfervers law no 

 liar, I have al'certaincd to be a globular 

 clufter, containing probably not lei's than 

 14,000 ftars." This clufcer mult have 

 a very powerful attxadtivc centre of gra- 

 vity, which miiy be able to keep many 

 far dillaut celollial bcjdies in controul. 

 An union of many fuch clulters will form 

 a ftill more powerful centre of gravita- 

 tion, whole influence may extend to a 

 whole region of fcattered liars. 



If a ftill more powerful, but more dit- 

 fufed exertion of attraction iliould be re- 

 quired than what may be found in the 

 union of cluli:ers, we have hundreds of 

 thoufands of liars, not to fay millions, 

 contained in very coinpreiicd parts of the 

 milky way. Thefe imuieale regions may 

 well occalion the lideieal motions refer- 

 red to ; and u liinilarity in the direttion 

 of thefo motions will want no iilulira- 

 tion. 



As additional rcafons for the admiirion 

 of far dillant centres of attratlion, as 

 ■well as projectile motiims in tl-.e liars tliat 

 are connetled with them, it may be ob- 

 fcrved, that independently of the folar 

 motions, the action of thefe caui'es will 

 be equally required to exiilaiu the ac- 

 knowledged motions of the Itars. For if 

 ihufuii be at rci'i, then Aiciurui umlt 



change its place more than 2" a year; 

 and confequently this and many otlier 

 Itai's, whicli are well known to change 

 their htuation, mult be fuppofed to liave 

 projectile motions, and to be fubjctt to 

 the attraction of far diliant centres. 



As the rel'ult of his feveral fpecula- 

 tions, Dr. llerlcliel oblerves, that *' it 

 oppears, in the piefunt hate of our J;dow- 

 ledge of the oblerved proper motions of 

 the liars, we have rcal'on to fix upon tlie 

 quantity of the folar motion t<j be fuck 

 a.s by an eye, [ilacetl at rii;lit angles tw 

 its direction, and at the diUance of .Si- 

 rius from us, ^YOuhl be fecn to defcribe 

 annually aa arc of 1". I1G992 of a tle- 

 grec, and its ^velocity, till we ;ire ac- 

 quainted with the real diliance of thi« 

 liar, can only be cx[)reil'ed by tlie pro- 

 portional numljer 111099,!. 



Tiie apparent velocities of Arcturus 

 anu x\ldebaraii, without a fohu- motion, 

 were fupp(jfed, by a table already rcibr- 

 rcd to, as 208 to 12 ; but when "the de- 

 ception ariling from its parallattic elfedt 

 is removed by calculation, thefe velocities 

 tn-e to each other only as 179 to 83, or 

 as 2 to 1 : and though Arcturus (iiU re- 

 mains a fiar that moves with great velo- 

 city, yet there are by t;he table four or 

 fue liars with nearly as much motion, 

 and four with more. This folar moti'«n 

 alio renio\es the deception by which the 

 motion of a ftar of the coufequciice ;'i 

 Ononis is lb concealed as hardly to lhe>» 

 any velocity ; whereas, by comjjutation, 

 we hnd tliat it really moves at a rate 

 which is fully equal to the motion of the 

 fun. 



The fimilarity of the directions of the 

 fidereal motions is an indication that the 

 liars, having fucli motions as well as the 

 lull, are acted upon by fome conucctmg 

 caufe, which Can only be attraction ; and 

 as attraction will not explain the oblerved . 

 phenomena without the exiftence of pro- 

 jectile motions, itmuft be admitted that 

 tile motions of the liars are governed by 

 the fame two ruling principles which re- 

 gulate the orbitual motions of the bodies 

 of the folar fyllom. It muft alfo be ad- 

 mitted that, we may invert the inference 

 from the operation of thefe caufes in our 

 fvlteni, conclude that their inilueiice ui;- 

 011 the hdeieal motions will tend to pro- 

 duce a limilar effect ; by which means 

 the probable motion of the fun, and of 

 the Itars in orbits, becomes a fubjcct ch.ot 

 may receive the afiiltance of argumenla 

 fupported by obfervation. 



At the lull iittin;j of this Society, a 

 paper 



