[ 5^8 ] 
as well froryi the obfervations made in one pofition of the 
inftmment as the other. If the inftrument had fufiered 
no change by being carried over the hill, that is to fay, 
if the line of collimation was not altered thereby, the re- 
fults fhould come out equally true from the obfervations 
taken in both portions of the inhrumcnt. On the con- 
trary, if the line of collimation fliould, by any means, 
have -fufFered any alteration between the obfervations 
m^'^^de at the two hations, this would caufe the dilFerence 
of latitude to appear too fmall, by the obfervations, made 
in one polition of the inftrument, by the quantity of the 
alteration, and as much too great, in the other polition 
of the initrument. But ftill the mean between the two 
refults, deduced from the obfervations taken in the two 
different pofitions of the inlfrument, would give the true 
difference of latitude ; and that equally, whether the line 
of collimation had buffered any change or not. There- 
fore this will be the beft method of comparing the ob- 
fervations together, and I fliall take a mean of all the re- 
fults, deduced from the obfervations taken in each pofi- 
tion of the inftrument feparately, and then a mean of 
thole means for the true difference of latitude. By fingie 
obfervations of ten Bars; viz. (3^ a, and n Cafliepese, and 
h ft 39? 45? 53 Dfaconis, made on both fides 
of the hill, with the plane of the fe6tor facing the Weft, 
after making the proper allowance for preceliion, aber- 
ration, and deviation, and femi-annual folar nutation of 
the earth’s axis (fee my tables annexed to my Obferva- 
tions made at the Ro} al Obfervatory), the apparent dif- 
ference 
