of the Equatorial Instrument. 
Table IV. is the natural secants to each degree, extracted 
from Sherwin's Tables ; being of such constant use in these 
computations, I have placed it with these tables of refraction. 
Table V. gives the correction of the time; viz. of the sun or 
star's distance from the meridian, in an observation with a 
portable equatorial, not previously adjusted to the meridian ; 
this quantity is to be multiplied also by the secant of the 
declination. 
Table VI. as its title imports, gives the correction of the 
meridian line in minutes and decimals, which was thought 
near enough for a portable instrument ; the quantity here 
found is to be multiplied by the secant of the altitude. 
Table VII. similar to Table I. only reduced to seconds of 
time, gives the refraction in right ascension, suited to the usual 
mode of dividing the small instrument ; viz. into civil hours 
and minutes. 
Table VIII. similar also to Table I. gives the refraction in 
declination. The arguments are the same in all the Tables ; 
viz. the altitude, and the angle of the horary and vertical circles ; 
which appeared to me the only means of making the Tables 
universal, and adapted to all latitudes.* 
Their foundation is this : 
* A table of refraction in right ascension and declination, for the latitude of Paris 
cnly, may be met with in the Connoissance des Temps, pour 1791. What I have given 
in the following account, viz. Tab. V. VI. VII. and VIII. I calculated for my own use, 
as early as the year 1774. 
Let AB be a portion of a 
vertical circle = the refraction 
in altitude ; DA a parallel to 
the horizon ; and D B a paral- 
MDCCXCIII. 
R 
