It appears plainly from the experiments which have been made on this subject, that the electric fluid is not able to pass through the glass; but yet it seems as if it was able to penetrate without much difficulty to a certain small depth, perhaps I might say an imperceptible depth within the glass; as Dr Franklin's analysis of the Leyden vial shews that its electricity is contained chiefly in the glass itself, and that the coating is not greatly over or undercharged.

It is well known that glass is not the only substance which can be charged in the manner of the Leyden vial; but that the same effect may be produced by any other body, which will not suffer the electricity to pass through it.

B

N

Fig. 20.
M

A

E

အစ်

129] *Hence the phænomena of the vial seem easily explicable by means of the 22nd proposition. For let ACGM, Fig. 20, represent a flat plate of glass or any other substance which will not suffer the electric fluid to pass through it, seen edgeways; and let BbdD, and EefF, or Bd and Ef, as I shall call them for shortness, be two plates of conducting matter of the same size, placed in contact with the glass opposite to each other; and let Bd be positively electrified; and let Ef communicate with the ground; and let the fluid be supposed either able to enter a little way into the glass, but not to pass through it, or unable to enter it at all; and if it is able to enter a little way into it, let bßdd, or bƐ, as I shall call it, represent that part of the glass into which the fluid can enter from the plate Bd, and ep, that which the fluid from Ef can enter. By the abovementioned proposition, if be, the thickness of the glass, is very small in respect of bd, the diameter of the plates, the quantity of redundant fluid forced into the space Bd, or B8, (that is, into the plate Bd, if the fluid is unable to penetrate at all into the glass, or into the plate Bd, and the space bo together, if the fluid is able to penetrate into the glass,)

R

G

* The following explication is strictly applicable only to that sort of Leyden vial, which consists of a flat plate of glass or other matter. It is evident, however, that the result must be nearly of the same kind, though the glass is made into the shape of a bottle as usual, or into any other form; but I propose to consider those sort of Leyden vials more particularly in a future paper.

will be many times greater than what would be forced into it by the same degree of electrification if it had been placed by itself; and the quantity of fluid driven out of Ep will be nearly equal to the redundant fluid in Bồ.

If a communication be now made between Bồ and Ep, by the canal NRS, the redundant fluid will run from BƐ to Ep; and if in its way it passes through the body of any animal, it will by the rapidity of its motion produce in it that sensation called a shock.

130] It appears from the 26th proposition, that if a body of any size was electrified in the same degree as the plate Bd, and a communication was made between that body and the ground, by a canal of the same length, breadth and thickness as NRS; that then the fluid in that canal would be impelled with the same force as that in NRS, supposing the fluid in both canals to be incompressible; and consequently, as the quantity of fluid to be moved, and the resistance to its motion is the same in both canals, the fluid should move with the same rapidity in both: and I see no reason to think that the case will be different, if the communication is made by canals of real fluid.

Therefore what was said in the beginning of this section, namely, that as great a shock would be produced by making a communication between the conductor and the ground, as between the two sides of the Leyden vial, by canals of the same length and same kind, seems a necessary consequence of this theory; as the quantity of fluid which passes through the canal is, by the supposition, the same in both; and there is the greatest reason to think, that the rapidity with which it passes will be nearly if not quite the same in both. I hope soon to be able to say whether this agrees with experiment as well as theory.

131] It may be worth observing, that the longer the canal NRS is, by which the communication is made, the less will be the rapidity with which the fluid moves along it; for the longer the canal is, the greater is the resistance to the motion of the fluid in it; whereas the force with which the whole quantity of fluid in it is impelled, is the same whatever be the length of the canal. Accordingly, it is found in melting small wires, by directing a shock through them, that the longer the wire the greater charge it requires to melt it.

132] As the fluid in BS is attracted with great force by the redundant matter in Ep, it is plain that if the fluid is able to penetrate at all into the glass, great part of the redundant fluid will be lodged in b8, and in like manner there will be a great deficience of fluid in ep. But in order to form some estimate of the proportion of the redundant fluid which will be lodged in bd, let the communication between Ef and the ground be taken away, as well as that by which Bd is electrified; and let so much fluid be taken from BS, as to make the redundant fluid therein equal to the deficient fluid in Ep. If we suppose that all the redundant fluid is collected in bồ, and all the deficient in ep, so as to leave Bd and Ef saturated; then, if the electric repulsion is inversely as the square of the distance, a particle of fluid placed anywhere in the plane bd, except near the extremities b and d, will be attracted with very near as much force by the redundant matter in ep, as it is repelled by the redundant fluid in b8; but if the repulsion is inversely as some higher power than the square, it will be repelled with much more force by bd, than it is attracted by ep, provided the depth bẞ is very small in respect of the thickness of the glass; and if the repulsion is inversely as some lower power than the square, it will be attracted with much more force by ep, than it is repelled by bd. Hence it follows, that if the depth to which the fluid can penetrate is very small in respect of the thickness of the glass, but yet is such that the quantity of fluid naturally contained in bd, or ep, is considerably more than the redundant fluid in B8; then, if the repulsion is inversely as the square of the distance, almost all the redundant fluid will be collected in bd, leaving the plate Bd not very much overcharged; and in like manner Ef will be not very much undercharged: if the repulsion is inversely as some higher power than the square, Bd will be very much overcharged, and Ef very much undercharged and if the repulsion is inversely as some lower power than the square, Bd will be very much undercharged, and Ef very much overcharged.

133] Suppose, now, the plate Bd to be separated from the plate of glass, still keeping it parallel thereto, and opposite to the same part of it that it before was applied to; and let the repulsion of the particles be inversely as some higher power of the distance than the square. When the plate is in contact with the glass, the

repulsion of the redundant fluid in that plate, on a particle in the plane bd, id est, the inner surface of the plate, must be equal to the excess of the repulsion of the redundant fluid in b8 on it, above the attraction of Ep on it; therefore, when the plate Bd is removed ever so small a distance from the glass, the repulsion of the redundant fluid in the plate, on a particle in the inner surface of that plate, will be greater than the excess of the repulsion of bd on it, above the attraction of Ep; for the repulsion of bd will be much more diminished by the removal, than the attraction of Ep: consequently, some fluid will fly from the plate to the glass, in the form of sparks: so that the plate will not be so much overcharged when removed from the glass, as it was when in contact with it. I should imagine, however, that it would still be considerably overcharged.

If one part of the plate is separated from the glass before the rest, as must necessarily be the case, if it consists of bending materials, I should guess it would be at least as much, if not more, overcharged, when separated, as if it is separated all at once.

In like manner, it should seem that the plate Eƒ will be considerably undercharged, when separated from the glass, but not so much so as when in contact with it.

From the same kind of reasoning I conclude, that if the repulsion is inversely as some lower power of the distance than the square, the plate Bd will be considerably undercharged, and Ef considerably overcharged, when separated from the glass, but not in so great a degree as when they are in contact with it.

:

134] § 7. There is an experiment of Mr. Wilke and Æpinus, related by Dr. Priestly, p. 258, called by them, electrifying a plate of air it consisted in placing two large boards of wood, covered with tin plates, parallel to each other, and at some inches asunder. If a communication was made between one of these and the ground, and the other was positively electrified, the former was undercharged; the boards strongly attracted each other; and, on making a communication between them, a shock was felt like that of the Leyden vial.

I am uncertain whether in this experiment the air contained between the two boards is very much overcharged on one side,

and very much undercharged on the other, as is the case with the plate of glass in the Leyden vial; or whether the case is, that the redundant or deficient fluid is lodged only in the two boards, and that the air between them serves only to prevent the electricity from running from one board to the other: but whichever of these is the case, the experiment is equally conformable to the theory*.

It must be observed, that a particle of fluid placed between the two plates is drawn towards the undercharged plate, with a force exceeding that with which it would be repelled from the overcharged plate, if it was electrified with the same force, the other plate being taken away, nearly in the ratio of twice the quantity of redundant fluid actually contained in the plate, to that which it would contain, if electrified with the same force by itself; so that, unless the plate is very weakly electrified, or their distance is very considerable, the fluid will be apt to fly from one to the other, in the form of sparks.

135] 8. Whenever any conducting body as A, communicating with the ground, is brought sufficiently near an overcharged body B, the electric fluid is apt to fly through the air from B to A, in the form of a spark: the way by which this is brought about seems to be this. The fluid placed anywhere between the two bodies, is repelled from B towards A, and will consequently move slowly through the air from one to the other: now it seems as if this motion increased the elasticity of the air, and made it rarer: this will enable the fluid to flow in a swifter current, which will still further increase the elasticity of the air, till at last it is so much rarified, as to form very little opposition to the motion of the electric fluid, upon which it flies in an uninterrupted mass from one body to the other.

In the same manner may the electric fluid pass from one body to another, in the form of a spark, if the first body communicates with the ground, and the other body is negatively electrified, or in any other case in which one body is strongly disposed to part with its electricity to the air, and the other is strongly disposed to receive it.

136] In like manner, when the electric fluid is made to pass [* See Articles 344, 345, 511, 516.]