A PENDULUM FOR THE REDUCTION TO A VACUUM. 
445 
Fifteenth set . — Results with the spheres at the centre of oscillation of the 
long rods. 
Diameter 
of the 
spheres. 
Brass rod. 
Steel rod. 
No. 
n 
Weight of 
adhesive 
air. 
Weight due 
to sphere 
alone. 
No. 
n 
Weight of 
adhesive 
air. 
Weight due 
to sphere 
alone. 
inches. 
0- oo* 
1- 46 
2- 06 
3-03 
43 
54 
55 
56 
4-083 
2-722 
2-186 
1-870 
1-484 
1- 749 
2- 352 
4-528 
0- 446 
1- 180 
3-382 
44 
57 
58 
59 
7-530 
2-248 
1-863 
1-774 
0-479 
0- 774 
1- 367 
3-719 
0- 405 
1- 039 
3-371 
These experiments confirm the results of the preceding- set, inasmuch as 
they show that, by fixing the spheres to this point of the rods also, the value of 
n is diminished : and there is moreover a greater regularity in the march of 
the values ; as the intervals of the coincidences were much more adapted for 
correct observation. They consequently furnish us with the means of deducing 
with a greater probability of accuracy, the quantity of air adhering to, or 
dragged by, each of the spheres independent of the rod. These values are 
given in the preceding Table, and have been deduced agreeably to the for- 
mula to which I shall presently allude, on the assumption that the weight of 
air dragged by the brass and steel rods, is accurately shown in the 13th set of 
experiments. The following Table exhibits in a different form the values above 
alluded to. 
Rods. 
Diameter of the spheres. 
1-46 
2-06 
3-03 
Brass 
0-446 
1-180 
3-382 
Steel 
0-405 
1-039 
3-371 
Mean — 
0-425 
1-109 
3-377 
The quantity of air dragged by the two separate portions of a rod (whether 
it be actually divided, as in the present case, or a portion of its influence on 
the circumambient atmosphere be interrupted and destroyed, as in the case of 
the discs in the 18th and 19th sets of experiments,) as well as the distance of 
* See the first note in page 443. 
MDCCCXXXII. 3 M 
