2 a Sir george shucks urgh’s Obfer vat ions 
Altitude from below at b. 
o / // 
Inclination of bc, — 1 1 20 26 
Correct for the part of the 1 ~ 
fignal obferved, — J 1 3 
Error of collimation, — o 59 
Correct for refradlion, — o 26 
Depreflion from above at c. 
o / // 
II 19 47 
Corre£t for the fignal, ■ — • — 59 
Error of collimation, — +59 
Effedt of refradlion, — +26 
True altitude of c from b, 1 1 1 7 23 
True deprefiion of e from c, 1 1 20 18 
Arch intercepted, or cur- J 2 iS 
vature, 
True altitude of c from b, 
deduced from the 
fervation at c. 
rom b, -j 
ie ob- f 
1118 o 
Mean of the two, or corrected altitude of c front b 
ii° 17' 41^'. 
Altitude at a. 
P 4 - 44 — 
o 
JL of inclination of abI 
the bafe, — v J 7 
Error of the line of col- 1 
_ J — ° 59 
limation. 
Corredl altitude of b from a, o 26 
Error of collimation. 
Depreflion at 
o / // 
o 27 4 
+ 0 59 
Corredl depreflion of a 1 « 
from b, — j 0 2 o 
Arch intercepted, — — o 27 
Altitude of b from a de-’j 
duced from the obfer- f O * 2 7 36 
vation at b, _ 
Mean of the two, or corrected altitude of b. from a 
= o° 26' 49" 
(h) It fhould feem from thefe two obfervations, that the error of the line of 
collimation had been afiumed too great; it has however, as I have before 
©bferved, nothing to do with the mean refalt: and this is, perhaps, one of the. 
bell means of difcovering the error of collimation, and the very method Mr,. 
de luc ufed, to adjull his level, though, as I have been informed by his brother, 
without taking into the account the effe£t of curvature, which, if his hori- 
zontal marks were 2000 feet diftant from each other, would amount to 20", and 
the error to half that quantity, 
s 
I have 
