[ i88 J 
heft deduced. To take a mean of the feveral re- 
sults will not be the proper method, except the 
feveral obfervations had been either all immerfions, - 
or all emerfions; but when both immerfions and 
emerfions have been obferved, then the beft method 
will be, to take a mean of all the refults given by the 
immerfions, and a mean of all the refults given by 
the emerfions, feparately ; the mean of thefe twcr 
means will be the true difference of the meridians,, 
and will be much more to be depended upon, than' 
if it was deduced from immerfions or emerfions 
only ; becaufe this method of comparifon removes 
the conftant errors arifing.from the differences of the- 
telefcopes, air, and eyes of the diftant obfervers ; 
thefe caufes of error affedting the feparate refults of 4 
immerfions and emerfions, in contrary ways. I fhall 
now only add the difference of meridians deduced 
from the obfervations contained in the preceding 
table, by taking, a proper mean -between the feveral 
ie£ult$o 
Meridian- 
