THE PACIFIC OCEAN. 
533 
were made only as an experiment, without aiming at much 1778. 
nicety. I alfo meafured moftly one way; whereas I ought t December ; 
to have brought alternately the reflected and diredl images 
to contrary lides, with refpecft to each other; reading the 
numbers off the quadrant, in one cafe, to the left of the be¬ 
ginning of the divifions; and, in the other cafe, to the right 
hand of the fame. It is evident, that half the fum of thefe 
two numbers muft be the true meafurement, independent 
of the error of the quadrant; and this is the method that I 
would recommend. 
But I am well affured, that it might have been obferved 
much nearer; and that this method may he ufeful when 
neither the beginning nor end of an eclipfe can he obferved, 
which may often happen. 
Immediately after the eclipfe was over, we obferved the 
diftance of each limb of the moon from Pollux and CLAriet.is ; 
the one being to the Eaft, and the other to the Weft. An 
opportunity to obferve, under all thefe circumftances, fel- 
dom happens; but when it does, it ought not to be omitted; 
as, in this cafe, the local errors to which thefe obfervations 
are liable, deftroy each other; which, in all other cafes, 
would require the obfervations of a whole moon. The fol¬ 
lowing are the refults of thefe obfervations: 
204° 22 7 07" 
Mr. King with j 1 
204 20 4 
204 27 45 
204 9 12 
mean 
mean 
nr . t 5 otArietis 
Myfelf with \ poUux 
ViArietis 
, Pollux 
Mean of the two means 
The time-keeper, at 4 h 30', to which time all 
the lunar obfervations are reduced 
i 
i 
204° 21' 5" 
204 18 29 
204 19 47 
204 04 45 
The 
