A PENDULUM FOR THE REDUCTION TO A VACUUM. 
421 
The results of all these pendulums agree very well together, except the 
brass one: and seem to show that in pendulums of equal length and of simi¬ 
lar construction, the factor for this additional correction depends on the form, 
and magnitude of the moving body ; and is not affected by its weight or spe¬ 
cific gravity. The mean of the whole makes n = T864*. I am unable to 
account for the discordancy of the brass sphere from the others ; unless it be 
in the determination of the specific gravity, which is certainly less than that 
of any brass I have yet examined: it being only 7'660 from a mean of three 
different weighings on three different days, and agreeing very well with each 
other. If the specific gravity be assumed equal to 7 ’ 8 or 7 ‘ 9 (which is still 
small,) the result of this pendulum would agree with the others: but I could 
never make it exceed 7‘67f- 
Second set .—Results with the 2-inch Spheres. 
On the Knife edge. 
On the Cylinder. 
5) Lead. 
6) Brass. 
7) Ivory. 
8) Lead. 
9) Ivory. 
fExp. 
n 
Exp. 
n 
Exp. 
n 
Exp. 
n 
Exp. 
n 
49—50 
1-764 
33—34 
1-736 
41—42 
1-752 
61—62 
1-811 
57—58 
1-760 
51—52 
1-732 
35—36 
1-732 
43—44 
1-759 
63—64 
1-682 
59—60 
1-722 
53—54 
T717 
37—38 
1-770 
45—46 
1-762 
55—56 
1-739 
39—40 
1-767 
47—48 
1-748 
Mean = 
1-738 
Mean = 
1-751 
Mean = 
1-755 
Mean = 
1-746 
Mean = 
1-741 
* There are some singular coincidences and discordancies in these results which though slight are 
worthy of notice. For instance, in the experiments with the lead sphere. No. 17—20 are almost 
identical; and so likewise are No. 21—24, yet differing from the former. Also in the experiments 
with the brass sphere, No. 9—12 are almost identical; and so likewise are No. 13—16, yet differing 
from the former. These and other cases of a like kind are trifling anomalies for which I cannot 
give any satisfactory explanation. 
f Some persons have supposed that if the ball be greased, the results might be affected : and if so, 
the present discordancy may have arisen from some accidental circumstance of this kind. It is pro¬ 
bable also that the results may vary according to the state of moisture or dryness of the atmosphere; 
or from some other unknown cause. On these points, there is certainly a wide field of inquiry open, 
but on which, at present, I have not leisure to enter. The true cause however, of the present dis¬ 
cordancy, I suspect to arise from some internal cavities in the sphere (indicated by the smallness of its 
specific gravity,) which are connected with the screw-hole, and thus suffer the escape of the included 
air when submitted to the action of the air pump. This contingency cannot be allowed for, in the 
computation; although it may be appreciable in the result. 
3 i 
MDCCCXXXII. 
