at the Island of Balta, and at Woolwich Common. 431 
Taking the ‘ rate” of the clock as determined day after day, 
by the sun or different stars, as I could observe them in succes- 
sion, it may be exhibited as follows: , 
1817. August Ist, 2d, rate from o Herculis +21-9 
August 2d, 3d, vee. ~ y Aquila 24-9 
August 3d, to the 9th, a Lyre 21-9 each. 
August Ist, 2d, Sr ekatn ‘Sun 21°95 
Do. 4th, 5th, tes Do. 21-9 
Do. 5th te the 10th, .. Do, 22:0 each. 
Dovwleh, Toth. oe. Do. 222 
Boe Tain, tothe, Do. 21-95 
Donlothe Tathy Oo as Do. 22°15 
Do. 14th, 15th, 2... Do. 22.1 
Dow15th, Lb6thy: cee xis a Lyre 22:0 
From these and some other observations from day to day, I 
inferred that no change either of temperature or of barometric 
pressure experienced here, occasioned a change of half a second 
in the “rate.” I therefore proceeded to deduce the average 
“rate ’? at Balta, from those observations on which | could most 
rely, on account of being enabled to see the star pass all the 
wires of the transit instrument. ‘The result is here presented. 
. a Ophiuchi, 13 observations, whole gain 287+1 mean 22:08 
eee ID ag ee alt 432°... 216 
oiplderculis, '2 0.555 oc eae a hc! Sea ep 
4 Serpentis, ©2 ...... cucees 1AM2) 9S °ODL-G 
a Lyre, BORAT. die%:,; 3 vee vee DOOM G2” |, CARDS 
yvAquile, 16 ....... ae eee e/@85030"... 21189 
aiAguile, (16 ~.....6 svegee “OSl66 '.. YZHORS 
B Aquile, 14 ...... bsepea! SCOR  . 2) 2RO7 
e Cygni, WS ote eet ade po WABZSS.°. 5) GOST 
94 2067°18 
Hence 2067:18~94=21":991, may be assumed for the mean 
rate (+) at Balta. 
This does not differ so much as four-tenths of a second, from 
any single rate furnished by the whole series of observations ; 
and is within an eighth of a second, of more than four-fifths of 
the daily rates, as ascertained by different stars. 
Both the nuts ab and cd c' d’, were below the bob, and close 
to each other during the Balta observations ; the respective in- 
_ dexes I and 1’, pointing to divisions agreed upon before the clock 
left Woolwich; viz. 1 to 13 and 1’ to 30. 
Notwithstanding the close agreement of the rates of the clock 
as determined by observations from day to day, it was exposed 
to considerable and rapid changes, especially of temperature, as 
will be seen from the table subjoined, 
Day 
