262 dr. Gregory's strictures on don Rodriguez. 



i 



or to a devia- not, at any of the stations, be out of the plane of the meridian, 

 plane "of 1 the* sna 'l select only two or three. As 1st, if the sector were 

 meridian; inclined to that plane, just so much would the path of any star, 

 in its apparent motion, be inclined to the horizontal wire of the 

 telescope j instead of which, both Colonel Mudge and Cap- 

 tain Colby assure me, that when a star came into contact with 

 the wire, the light of the star would appear on both sides of 

 the wire for about three-fourths of a minute of time, the light 

 on each side being equal at the central wire : which of itself is a 

 positive proof. But, 2dly. had the sector been out of the plane of 

 the meridian, the times of the transits of the extreme stars em- 

 ployed, as compared with two excellent time-keepers, must 

 have shewn it. Farther, the errors arising from a wrong plane 

 of the meridian, being comparatively very great in the extreme 

 stars, and small in those near the zenith, it would follow that 

 the error in Capella, which is almost at the extremity of the 

 arch, would be great, compared with those in /3. Draconis, 

 k Cygni, &c. which were within a small distance of the ze- 

 nith. But the amplitude of the arch, between Dunnose and 

 Arbury Hill, as derived from Capella, is 1° 36' 20* ''02, 

 while those derived from the other two stars, are J °36" 19-"42, 

 and 1° 36' 1Q*"Q4: a coincidence which proves that the in- 

 strument could not possibly have any perceptible deviation from 

 the plane of the meridian at either station. Other reasons for 

 coming lo the same conclusion will appear, on attending to the 

 precautions in adjusting by double azimuths, &c. as described 

 in the Phil. Transactions. 



The correct position of the sector in all respects is further 

 or to its use proved from this : that the observations, however distant in 

 m anyway, point of time, when the proper corrections for aberration, nu- 

 tation, &c. are applied to them, reduce always very nearly .to 

 the same mean place. 



Hence, it must be obvious, that no error could arise, as Don 



Rodriguez suspects, from the instrument, whether in " vertical 



position, construction, or some accidental derangement." I 



shall now advance still farther, and prove that there is no 



rp. error in fact. For if there were any error in the zenith dis- 



not be an error tances at Arbury Hill, it would at once be detected on com- 



even of half a par j son -with the observations at Blenheim. Now, the distance 



Arbury or between the parallels of latitude of Blenheim and Arbury, 



139,822 



