[ 367 ] 
Reduce the parts of 96 to degrees, and take the mean 
between the zenith diftances fliewn on each fcale, and 
the zenith diftance of ^ Draconis on the quadrantal arch 
/ // / // 
will be 9 21,7, and oil the arch of excefs 8 54, whence 
/ n 
the true zenith diftance is 9 7,8, and the error of the 
// 
line of collimation 1 3,8, to be fubtradled from the num- 
bers fliown on the limb of the quadrant. In like man- 
ner we fliall find the true zenith diftance of y Draconis 
0 / // // 
1 6 1 9,8, and the error of the line of collimation 13,5. 
If we fuppofe the apparent declination of /3 Draconis on 
o / H 
July 1 2th to be 52 28 52,3, that of y Draconis 
o f it 
51 31 41,7, we have the latitude from the former 
52° 38', and from the latter 52° 38' 
N. B. Some obfervations on thefe two ftars in July 
iJTii give the fame latitude within lefs than but 
make the error of the line of collimation 23'^ to be fub- 
tradted. I fufpedl the line of collimation is liable to 
fmall variations in portable quadrants, if not in all. 
Zemt^i 
VoL. LXV. 
D d d 
