tte GEORGIUM SIDUS. 315 



had not yet arrived at its perihelion. Hence it is demonflrable, 

 that the differences of the arches defcribed in equal times 

 fhould form a feries of numbers continually decreafing, very 

 flowly at firft, but afterwards more rapidly. 



UPON this principle, we may venture to correct the obferva- 

 tions. In this correction there is ftill a choice ; for we may 

 make the decreafe of the feries either more or lefs rapid. The 

 ellipfes which arife from the extremes of the feries formed up- 

 on this principle, will evidently be the limits which compre- 

 hend the principal elements of the eccentric motion ; and, 

 fince we allow ourfelves very little liberty in the correction, it 

 is prefumable, that thefe limits will not be very wide. 



FROM the above obfervation of the Planet in its ftationary 

 point, we find that its angular velocity does not greatly exceed 

 that of a Planet revolving in a circle j and a fimilar ufe being 

 made of Mr MASKELYNE'S firft obfervations, will mow, that 

 the heliocentric motion of the Planet in April 1781 hardly ex- 

 ceeded the motion in a circle at the fame diftance. We may, 

 therefore, prefume that its true anomaly does not much exceed 

 90. Therefore, the feries of firft differences, adapted to this 

 fituation, muft decreafe very flowly, whilft the fecond differen- 

 ces muft increafe alfo very flowly. This will appear by exa- 

 mining the tables of any of the Planets. I fhall, there- 

 fore, begin by giving to the fecond differences a very fmall 

 increafe, and to the firft differences a very fmall diminu- 

 tion. This will be done by a correction not exceeding 3" in 

 any of the obfervations ; and this muft be allowed to be far 

 within the limits of probability. The firft obfervation has its 

 longitude diminiflied i" ; the fecond has its longitude increafed 

 2-7" ; the third has its longitude increafed by the fame quanti- 

 ty, and the fourth and fifth have their longitudes increafed 3". 

 The times correfponding to the above mentioned equal inter- 

 vals, and the correfponding corrected longitudes, cleared from 



R r 2 the 



