230 PHILOSOPHICAL TRANSACTIONS. [aNNO 1787- 



of latitude of the two observatories, which, from what has been said of the ob- 

 servations on which the respective latitudes were founded, cannot be supposed to 

 differ above 3 or 4" from the truth. What then becomes of the uncertainty of 

 \5" supposed by the late M. Cassini ? 



The same difference of latitude I find nearly from a comparison of my own ob- 

 servations of y and |3 Draconis, taken with the zenith sector in 1768, with those 

 of the Abbe de la Caille in 1750 and 1756, given in his Fundamenta Astronomiae, 

 after making the proper allowances for aberration, precession, and nutation, and 

 correcting my observations by Dr. Bradley's refraction, and the Abbe de la 

 ■ Caille's by his table, and making allowance for the distance of the Abbe de la 

 Caille's Observatory from their Royal Observatory; viz. 2,^38' 25" A from y Dra- 

 conis, and 2° 38' 26^'. 1 from j3 Draconis ; the mean being 2° 38' 25''.7, differing 

 only 0".3 from that stated above; but from Dr. Bradley's observations 2° 38' 2A"g, 

 and 2° 38' 27^'.2, mean 2° 38' 2& O. It is too well known to astronomers to 

 need my pointing out, that the best method of determining the difference of 

 latitude of places, differing but little in latitude, is by such differences of zenith 

 distances of stars passing near the zeniths, as the two above cited, observed at 

 both places, in the same manner as the amplitude of the celestial arc is observed 

 for finding the length of a degree of the meridian by comparison with geometrical 

 measures. 



The question now will be, on what foundation was the late M. Cassini's sup- 

 position of an uncertainty of 15'' in the latitude of Greenwich built? This ap- 

 pears evidently to have been on a passage in the Abbe de la Caille's researches 

 into the astronomical refractions and latitude of Paris, contained in the Memoires 

 of the Royal Academy of Sciences for 1755, p. 578, 579, where M. de la Caille 

 takes the differences of zenith distances of 14 stars observed by Dr. Bradley (in 

 correspondence to the same observed by himself at the Cape of Good Hope, for 

 determining the moon's parallax in declination) published in the Memoires of the 

 Royal Academy of Sciences for 1752, and the same observed by himself at Paris, 

 after his return from the Cape, and correcting them for the difference of the re- 

 fractions at the respective zenith distances, according to his own table of refrac- 

 tions, and the known apparent motions of the stars, finds the mean 2° 37' 23''''.9, 

 which added to 48° 51' 29''.3, his latitude at the College of Mazarine, gave him 

 51° 28' 53".2 for the latitude of Greenwich, exceeding Dr. Bradley's latitude by 

 13 or 14''. 



Now the legitimacy of this conclusion depends on a supposition that both in- 

 struments measured the true angle, or that their total arcs were justly laid off, and 

 that the Abbe de la Caille's table of refractions is just. The first indeed has been 

 proved with respect to Dr. Bradley's quadrant, but never has been attempted with 

 respect to the Abbe de la Caille's sextant ; for the examination which the Abbe 



