VOL. LXIII.] VMILOSOPHICAL TRANSACTIONS. 3^5 



As the results of the observations only, and not the observations themselves, 

 are communicated, Mr. H. observes, that there is a very considerable difference 

 between the conclusions of the two astronomers for the same year 1763, and 

 declares his suspicion, that if the apparent (for such he apprehends them to be) 

 were reduced to the mean distances, they would probably afford a confirma- 

 tion of the diminution of the ecliptic. For the following observations of the 

 sun's zenith distance, made at Shirburn castle, near the summer solstices of the 

 years 1743, 1/46, 1748, and 1766, and of Arcturus in the years 1743, 1746, and 

 1766, when reduced to their mean state at the solstice, do not confirm the 

 assertion of Mr. Cassini, but are an evident and absolute proof that the 

 obliquity of the ecliptic has sensibly diminished during an interval of 23, and 

 even of 1 8 years. 



The observations of 1743 were made with a mural quadrant of 5 French feet, 

 constructed by the late Mr. Sisson: but as the linear divisions were found to be 

 somewhat less accurate than was expected, and as the body of the quadrant was 

 not framed with proper strength and solidity, Mr. Bird was employed in the 

 summer of the year 1745, by the Earl of Macclesfield, (the body of the instru- 

 ment having been strengthened by screwing a large and broad plate of brass on 

 the cross bars), to put a set of points on the limb between the QO and 96 arches 

 of linear divisions. By these operations the line of collimation was found to 

 have varied, and to be s= 6".3, by which the zenith distances were given too 

 small, by the positive divisions, from the end of 1746 to the end of June 1751, 

 when Mr. Bird bisected the spaces between the points which he had formerly 

 added in 1745. But after the year 1751, the error of the line of collimation 

 was := 2". 6, as appears from observations of y Persei, (3 and y Draconis, by 

 which the zenith distances are also given too small; and in that state the instru- 

 ment continued to the year 1767, when a new set of wires was put into the 

 telescope, and the line of collimation thereby altered. The error of the line of 

 collimation from 1743 to 1745 cannot directly be ascertained, for want of zenith 

 observations; but, from some indirect methods, it should seem that the error 

 was as nearly as possible = 2", to be added to the observed zenith distances. 



Thus by a series of observations of the sun's zenith distances, from the 7th to 

 the 27th of June, 1743, when corrected for his semidiameter and refraction, the 

 medium of all the 12 days, when reduced all to the solstice, is as follows: 



A similar set of 14 days observations, from 



May 31, to June 30, 1746, give » 



TlieMean 28° 10' 58. C;" For the Mean 28" 10 52.5"'' 



Sun's parallax — 4. 1 Sun's parallax — +. 1 



28 10 54.1 28 10 48.4 



Nutation +6.7 Nutation -(-9.4 



28 11 0.8 28 10 57.8 



Line of collimation +2. Error of tlie line of coUination +6.3 



Mean solstitial zenith dist., 1743 28 11 2.8 Meansolstitial zenith dist., 174t) 28 11 4.1 



3 E 2 



