472 
MR. BAILY ON THE CORRECTION OF 
of the seconds pendulum. In all such cases, however, it is presumed that the 
two knife edges are adjusted very nearly to synchronism. If we apply these 
principles to Captain Sabine’s results, we shall have the following values for 
the number of vibrations if the pendulum were rendered perfectly synchronous. 
Slider. 
A. 
B. 
(A-B) 
If synchronous. 
1-500 
1-566 
1-633 
86069-00 
69-04 
69-10 
86070-26 
69-61 
69-14 
1 1 1 
o o ^ 
o cti vb 
86067-72 
68- 46 
69- 06 
There is a difference, in these three values, of 6 and 7 tenths of a vibration ; 
and if one is to be preferred to the other, it should be that which is the result 
of the greatest number of experiments, which appears to be the second value 
here given. But, they all want the requisite corroboration of the knife edge B. 
Method of observing and of reducing the Observations . 
Before I conclude this paper, it may be proper to say a few words on the 
method employed in making the experiments above alluded to, and of the data 
used in the reduction of the observations, in order that the circumstances, 
under which each experiment has been made, may fully appear, and that each 
step of the computations may be verified at pleasure. 
The clock used for observing the coincidences is an excellent one made by 
Molyneux, having a mercurial pendulum, with a long tail piece, furnished 
with two circular segments of gilt paper, which reflect a very brilliant light: 
the distance between these segments is variable at pleasure, in order to suit the 
size of the different pendulums under experiment. The rate of the clock is 
ascertained by a daily comparison with another clock (made by Hardy) regu¬ 
lated to sidereal time ; the rate of which is determined by means of a 30-incli 
transit instrument. Both these clocks go very well; and with respect to the 
experiments detailed in this paper, which are merely comparative, do not afford 
the source of any appreciable error. The clock used with the experiments, and 
which I shall, for the sake of distinction, call the Pendulum-clock, has been in 
all the cases, regulated to mean solar time; except when used with the long 
cylindrical rod (No. 21) and with the long brass tube (No. 35—38) ; where it 
was necessary to alter the length of the clock pendulum in order to obtain con- 
