the outside, that is, in which the thickness of the glass is almost equal to the outside semidiameter, than it is in the first mentioned. cylinder.

Nothing certain, however, can be inferred as to this point, as in all probability the four pieces of flint glass used in these experiments and the two flat pieces used in [Art. 370] did not consist exactly of the same kind of glass, as indeed appears from their specific gravities.

385] The three green cylinders, indeed, were all made at the same time and out of the same pot, so that it seems difficult to suppose that there should be any difference of that kind between them*. But then I had no flat plates to compare them with.

On the whole, I think we may with tolerable certainty infer that the ratio of the real to the computed charge is not very different from what it is in flat plates, whatever is the proportion which the thickness of the glass bears to the diameter of the cylinder, though it seems to be not exactly the same.

* Though it seems not likely that there should be any difference in the nature of the glass of which the three green cylinders consisted, yet I am not sure that there was not, for the inside of the glass, that is, that part which was nearest to the inside surface, was manifestly more opaque and of a different colour from the outside, and the separation between these two sorts of glass appeared well defined, so that the cylinder seemed to consist of two different coats of glass lying one over the other. The distinction was the most visible in those cylinders which consisted of the thickest glass and in the thickest part of those cylinders. The specific gravities, however, do not indicate any difference in the nature of the glass. What was the reason of the above-mentioned appearance I cannot tell.

WHETHER THE FORCE WITH WHICH TWO BODIES

REPEL IS AS THE SQUARE OF THE REDUNDANT
FLUID, TRIED BY STRAW ELECTROMETERS *.

386] If two bodies, A and B, placed near to each other, are both connected to the same overcharged Leyden jar, and the force with which this jar is electrified is varied, everything else remaining unaltered, the force with which A and B repel each other ought by the theory to be as the square of the quantity of redundant fluid in the jar, supposing the distance of the bodies A and B to remain unaltered. For the quantity of redundant fluid in A is directly as the quantity of redundant fluid in the jar, and therefore the force with which each particle of redundant fluid in B is repelled by A is also directly as the quantity of redundant fluid in the jar, and therefore as the number of particles of redundant fluid in B is also as the quantity of redundant fluid in the jar, the force with which B is repelled by A is as the square of the quantity of redundant fluid in the jar.

387] In order to try whether this was the case, I made use of the following apparatus †.

CD (Fig. 31) is a wooden rod 43 inches long, covered with tinfoil and supported horizontally by non-conductors. At the end

[Title supplied from Cavendishes Index to his experiments, Art. 563.]

+[Arts. 563, 567, also Art. 525.]

C is suspended, as in the figure, the electrometer described in Art. 249, and at the other end D is suspended a similar electrometer, only the straws reached to the bottom of the cork balls A and B, but not beyond them, and were left open so as to put in pieces of wire, and thereby increase their weight and the force with which they endeavoured to close. The lower ends of these wires when used were just even with the bottom of the cork balls, and were kept in that situation by wax, the wax being cut off even with the bottom of the corks, so as to leave no roughnesses to carry off the electricity. In like manner, when the wires were not used, the ends of the straws were closed up with wax.

388] The proportion which the force with which the balls of this electrometer endeavoured to close when the wires were inserted bore to that with which they endeavoured to close without the wires was thus found. The weight of the straw

A

B

with its

(7.6

6.65

ball and centre pin but without its wire was found to be grains, and the distance of its center of gravity from the center of (5.36 suspension was 5.285

inches, as was found by balancing it on the

edge of a knife. Consequently the force with which this straw when put in its place, endeavours to descend towards the perpendicular, supposing it to be removed to a given distance from it, was (7.6 × 5·36

as

6.65 x 5.285*

The weight of the wire inserted was

(12.05
10

grains, and half its

(1.23 length was inches, so that as the distance of the bottom of 1.00 the cork balls from the center of suspension was 111 inches, the distance of its center of gravity from the center of suspension was 9.87

inches, and therefore the excess of the force with which the 10.1 ball endeavours to descend towards the perpendicular when the wire is inserted above that with which it endeavours to descend without [the wire] is to the force with which it endeavours to descend without the wire as (1203 × 9.87

10

× 10.1

to

(7.6 × 5:36

(6·65 × 5.285'

or as

(2.92

12.88

to one.

Therefore the force with which the electrometer endeavours to close when the wires are inserted is to that with which it endeavours to close without the wires as 3·9 to 1.

389] E and F are two coated Leyden vials, nearly of the same size. The outside coatings of both communicate with the ground, and the inside coating of E communicates with CD, but not that of F.

390] The way in which I tried the experiment was as follows. I first compared the electrometer C with the electrometer D without the wires, and found that when the jar E was electrified to such a degree as to make D separate divisions, C separated

(13

(12

13 divisions, so that the same degree of electrification which made C' separate {12 divisions made D separate (13 divisions.

I then put the wires into the electrometer D, and put the larger of the two vials in the place of E, and electrified E and consequently the rod CD and the two electrometers till D sepa

The wire by which E was electrified was then immediately taken away and a communication made between E and F, so that the redundant fluid in E and CD and the electrometers was communicated to F.

(157

It was found that the electrometer C then separated 14

divisions.

The experiment was then repeated in the same manner, except that the smaller vial was placed at E. It was found that if

(13
12

E was electrified till D separated divisions, then on making a

communication between E and F, C separated

(131

121

divisions.

391] From hence we may conclude that if the vials had been

exactly equal and E had been electrified till D separated

(13 12

divisions, then on making a communication between E and F, C

(141 131

would have separated divisions.

But it appears from the first mentioned part of the experiment, that the same degree of electrification which makes C separate

(14) divisions is sufficient to make D without the wires separate

131

(13

divisions. From whence it appears that if the jars are exactly

(11종 equal, and one of them is electrified till the electrometer D with (13 the wires separates divisions, and its electricity is then com12

municated to the other vial, the electricity will be of that degree of strength which is necessary to make the same electrometer

(13

71177

without the wires separate 1 divisions, that is, very nearly the same as before, or as it did with the wire before the communication of the electricity.

But if the vials are equal, the quantity of redundant fluid in the first vial, after its electricity is communicated to the second, will be very little more than half of what it was before the communication, for the quantity of redundant fluid in the rod DC and the electrometers is trifling in comparison of that in the vial*, and consequently it appears that the distance to which the electrometer with the wires in it separates with a given quantity of redundant fluid in the vial is very nearly the same as that to which it separates without the wires when there is only half that quantity of redundant fluid in the vial.

Therefore as the force with which the electrometer endeavours to close by its weight when the wires are in is to that with which it endeavours to close without the wires as 39 to 1, it appears that the force with which the balls of the electrometer are repelled with a given quantity of redundant fluid in the vial, is to that with which they are repelled when there is only half that quantity of redundant fluid in the vial as 39 to 1 (supposing the distance

[In a sentence which Cavendish has scored out in his MS. we read-] The charge of the two vials together was found to be 2168 inches. The diameter of the rod CD was at a medium about of an inch. [This would make the computed charge of the rod 9.7 inches.-ED.]