39 



result of a long series of " good turns," every one of which has 

 tended to help forward its successors. 



The sun's rotation may be accounted for in the same way. 

 All the planets and all their satellites act upon him as well as he 

 acts upon them ; their combined power altogether is not suflfi- 

 cient to move him from his place, but it may be quite sufficient 

 to pull him round on his axis. The planets all revolve in the 

 same direction ; and their pulling power acting, as it has acted 

 for countless ages, may have resulted at last in a mean movement 

 of rotation, just like that of a fly wheel, which moves with regu- 

 lar velocity notwithstanding the irregular application of the mov- 

 ing power. The planets may be considered as so many men or 

 horses pulling by so many ropes at the circumference of a mon- 

 strous wheel, which, when once set fairly in motion, continues to 

 move of itself, with a velocity proportional to, as well as produced 

 by, the various efforts of the combined powers. And, perhaps, 

 if the sun himself be only the satellite of some mightier centre, 

 his rotation will have to be ascribed in part to that also,\as well 

 as the earth's to the action of the sun. 



The error respecting the moon's alleged rotation seems to have 

 occasioned, or served at least to confirm another error respecting 

 the measure of rotation^ and the equation of time, "The length 

 of the astronomical mean day, which is divided into 24 hours, is 

 determined by the interval which elapses between two consecu- 

 tive passages of the sun over the meridian of the same place, 

 supposing this apparent motion of the sun to be performed with 

 an uniform velocity. But it is necessary to observe, that our 

 earth does not occupy quite 24 hours in its rotation, on account 

 that in the same time, which it employs in revolving round its 

 axis, it advances in its orbit towards the east about a degree in 

 space, corresponding to 4 minutes, or more exactly to 3 minutes 

 5Q seconds of time. Hence it follows, that the interval between 

 two passages of a fixed star over the same meridian, which mea- 

 sures the true time of the eartKs rotation^ or of the sidereal day, 

 is only 23 hours, 6Q minutes, 4 seconds. The sidereal day is 

 the period in which the earth makes one complete revolution on 

 her axis^ and is measured by the interval between tioo successive 

 transits of any fixed star over the same meridian,^ 



The slightest consideration should suffice to show the absurdity 

 of such a notion. Suppose a body to revolve in a perfectly cir- 

 cular orbit ; and its own circumference to be of exactly the same 

 dimensions as the length of one degree of the orbit described by 

 its innermost point of circumference. As already remarked, 

 every revolving body has a certain definite breadth or thickness, 



