236 PHILOSOPHICAL TRANSACTIONS. [aNNO 1787. 



Stars and the obliquity of the ecliptic found by both, will be very near the truth, 

 independent of the justness of the total arcs, though their respective refractions 

 may be suitable only to their own particular instruments. But, for the reasons 

 before given, I apprehend the Abbe de la Caille's refractions to be much too 

 large, and Dr. Bradley's to be very near the truth. 



I shall now close my inquiry into the latitudes of Greenwich and Paris, and 

 Dr. Bradley's and the Abbe de la Caille's refractions, by a remark naturally 

 arising from my comparison of, and 'endeavours to reconcile, their observations, 

 which I desire to submit to the consideration of astronomers, it not having, 

 that I know of, been made before; that a table of refractions should be made 

 for every vertical instrument from observations made with itself turned alter- 

 nately north and south; and that the table, so made, applied to observations 

 made with it, will give the true zenith distances, whether the total arc of the 

 instrument be accurately just, or affected with a small error, or however un- 

 equally it be divided below the pole, provided the divisions are equal between 

 themselves in the part of the instrument lying between the equator, the zenith, 

 and the pole. 



It remains to give some account of the longitude of Greenwich, or rather of 

 the difference of meridians of Greenwich and Paris, in reply to the late M. 

 Cassini's doubts on the subject. This had been settled by Dr. Bradley at Q*" 20% 

 as he informed me himself, and that he had deduced it from eclipses of Jupiter's 

 first satellite observed at both places, and that he had found it come out the same 

 both from the immersions and emersions. This quantity had been inserted in 

 the table of latitudes and longitudes of places, prefixed to Dr. Halley's tables, on 

 the authority of Dr. Bradley, so long ago as the year 1749, the date of the pub- 

 lication of those tables, and was generally admitted by astronomers till the year 

 1763, when the late Mr. James Short, f. r. s., computed it from the 4 transits 

 of Mercury over the sun in 1723, 1736, 1743, and 1753, observed at Paris, 

 London, and Greenwich, to be 9™ 16*. See Philos. Trans., vol. 53, p. 158. 

 In the year 177^, I requested the late Mr. Wargentin, the learned secretary of 

 the Royal Academy of Sciences at Stockholm, and author of the improved tables 

 for computing the eclipses of Jupiter's satellites, who collected observations of 

 them from the principal observatories of Europe, in order to the further improve- 

 ment of the tables, to inform me what difference of meridians of Greenwich 

 and Paris resulted from my last 1 years observations of the eclipses of the first 

 satellite of Jupiter, compared with those made by M. Messier at Paris. In the 

 answer which he favoured me with, inserted in the Philos. Trans., vol. 67, p. 

 162, he set down the result of the comparison of 8 corresponding immersions 

 and 9 emersions observed on both parts, by myself and M. Messier, from which 

 he deduced the difference of meridians of the Royal Observatories of Greenwich 

 and Paris 9"^ 35S By 2 corresponding immersions and 9 corresponding emer- 



