A PENDULUM FOR THE REDUCTION TO A VACUUM. 
417 
state that the knife edges A and C are rendered synchronous, or nearly so; 
and that B and D are also rendered synchronous, or nearly so. It follows 
therefore that each pair (when properly reduced,) should give the same result 
for the length of the simple pendulum. But the discordancies which they ex¬ 
hibit have been already described by me in the work just quoted, and have 
given rise to three separate papers on the subject by Captain Everest, Mr. 
Gompertz and Mr. Lubbock # . The specific gravity of the pendulum, deduced 
from a piece of metal said to be from the same casting, I found to be 8‘060 : 
its weight is 231437 grains. 
No. 35, 36, 37, 38 are four of the knife edges, or rather planes, A, C, 
a, c, of a brass cylindrical tube, or rather tubes, for it is formed of 7 different 
tubes drawn closely one within the other ; so that their joint thickness, which 
is very firm and compact, and appears as one solid body, is about O’13 inch. 
The diameter is 1^ inch on the outside, and it is 56 inches long: the 
ends are not closed. The specific gravity of the metal I found, by 
weighing a piece of the tube itself, to be equal to 8 - 406 : but as the 
included air must be taken into account, the diminished specific 
gravity of the moving body (deduced agreeably to what is stated in 
page 411,) will be 3'034. Its weight is 81047 grains. This pendulum 
is of a totally different construction from any hitherto made: for 
instead of being fitted up with steel knife edges that vibrate on agate 
planes, it is furnished with steel planes that vibrate on a pair of agate 
knife edges which is common to all the planes. The mode of suspen¬ 
sion therefore is, in this case, reversed. The pendulum has six planes: 
but, as two of them (B and b ) have not yet been used, I shall confine 
my remarks to the four above enumerated. At the distance of 4 inches 
from each end of the tube is placed one of the planes, fastened to a 
brass collar, firmly fixed to the tube. At 12 inches distance from each of these, 
towards the centre, is placed another plane ; thus forming four in the whole. 
* In this last-mentioned paper, which is inserted in the Phil. Trans, for 1830, page 201, Mr. Lub¬ 
bock has shown the effect on the number of vibrations of a given pendulum, corresponding to given 
deviations in the position of the knife edges. And the result is, that no error of any considerable (or 
even appretiable) magnitude can arise from such causes, when the artist uses even the most ordinary 
precaution in fixing the knife edges in their proper position. Tire discordancies, I believe, arise from 
irregularities in the knife edge or planes ; as I shall more particularly allude to, in the sequel. 
