438 Mr. South on the discordances between the 
various determinations of 1821 and 1822, exhibited in Tables 
V. and VI. it will be seen, although the difference is not con- 
stant, yet that within two or three days, its amount does not 
greatly vary ; by collecting therefore consecutive transits, in 
pairs, each of which shall always contain a result, derived 
from observation made during exposure of the instrument, we 
may probably arrive at some conclusion, which, although not 
demonstrative, will still merit considerable confidence. Let us 
begin with 1821. 
From Table V. 1821. 
i 
Instrument exposed. 
Instrument defended. 
Difference. 
seconds. 
seconds. 
seconds. 
August 22 
+ 0-755 
August 21 
+ o-733 
+ 0.022 
Sept. 3 
+ 0.672 
Sept. 2 
4- 0.661 
+ 0.011 
— — — 3 
+ 0.672 
4 
4. 0.661 
+ 0.01 1 
— — 5 
4- 0.660 
— 4 
4- 0.661 
— 0.001 
“ 2 5 
+ 0.701 
— 24 
+ 0.773 
— 0.072 
1 
Mean diff. of the 5 pairs 2= 
1 
0 
b 
0 
01 
00 
From Table VL 1822. 
j Instrument exposed. 
i 
Instrument defended. 
Difference. 
| 
seconds. 
seconds. 
seconds. 
J March 
1 
+ O.030 
February 28 
+ 0.225 
— 0.195 
May 
21 
+ O.861 
May 
22 
+ O.932 
— 0,071 
3 1 
+ O.971 
June 
1 
+ O.826 
+ O.I45 
June 
2 
+ 0.826 
June 
1 
+ O.826 
+ 0.000 
— 
2 
+ O.826 
June 
3 
+ O.873 
— 0.047 ! 
4 
+ 0.704 
June 
3 
+ O.873 
— 0.169 
7 
+ 0.705 
June 
6 
+ O.927 
■ 0.222 
Decem. 
22 
+ 0.164 
Decim. 
23 
+ O.I58 
+ 0.006 
Mean diff. of t 
ie 8 pairs = 
— O.069I 
Hence, in 1821, the mean of 5 observations, obtained when 
the instrument was exposed to the sun’s rays, varies from 
