have broke while I was charging them, have shown, besides the perforation in the body, a trace on both sides of the neck, where the polish of the glass was taken off the breadth of a straw; which proved that great part at least of the charge, probably all, had passed over that trace. I was once present at the discharge of a battery containing thirty jars, of which eight were perforated and spoilt at the time of the discharge; yet the effect of the charge on the bodies upon which it was intended to operate, did not appear to be diminished. Another time I was present when twelve out of twenty jars were broken at the time of the discharge; yet the effect of the charge, which passed in the regular circuit, was the same as it would have been if they had remained whole. Were those perforations an effect of the charge within the jar forcing itself through the glass to get at the outside, other difficulties would arise and demand explanation. 1. How it happens, that in eight bottles, and in twelve, the strength to bear a strong charge should be so equal, that no one of them would break before the rest, and thereby save his fellows; but all should burst at the same instant. 2. How it happens, that they bear the force of the great charge till the instant that an easier means of discharge is offered them, which they make use of, and yet the fluid breaks through at the same time.

My conjecture is, that there has been, in the place where the rupture happens, some defect in the glass, some grain of sand perhaps, or some little bubble in the substance nearly void, where, during the charging of the jar, the electric fluid is forced in and confined till the pressure is suddenly taken off by the discharge, when, not being able to escape so quickly, it bursts its way out by its elastic force. Hence all the ruptures

happen nearly at the same instant with the regular discharge, though really a little posterior, not being themselves discharges, but the effects of a discharge which passed in another channel.

QUESTION II.

When a strong explosion is directed through a pack of cards or a book, having a piece of tinfoil between several of its leaves, the electrical flash makes an impression in some of those metallic leaves, by which it seems as if the direction of the electric explosion had gone from the outside towards the inside, when, on the other metallic leaves, the impression is in such a direction, that it indicates the current of electrical fire to have made its way from the inside of the phial towards the outside; so that it appears to some electricians, that, in the time of the explosion of an electrical phial, two streams of electrical fire rush at the same time from both surfaces, and meet or cross one another.

ANSWER.

These impressions are not effects of a moving body, striking with force in the direction of its motion; they are made by the burs rising in the neighbouring perforated cards, which rise accidentally, sometimes on one side of a card, and sometimes on the other, in consequence of certain circumstances in the form of their substances or situations. In a single card, supported without touching others, while perforated by the passing fluid, the bur generally rises on both sides, as I once showed to Mr. Symmer at his house. I imagine that the hole is made by a fine thread of electric fluid first passing, and augmented to a bigger thread at the time of the explosion, which, obliging the parts of a card to recede every way, condenses a part within the

substance, and forces a part out on each side, because there is least resistance.

QUESTION III.

When a flash of lightning happens to hit a flat piece of metal, the metal has sometimes been pierced with several holes, whose edges were turned some the one way and some the other; so that it has appeared to some philosophers, that several streams of electrical fire had rushed in one way, and some the opposite way. Such an effect of lightning has been published lately by Father Barletţi.

ANSWER.

This will be answered in my remarks on Mr. Barletti's book; which remarks, when finished, I will send you.

QUESTION IV.

Though, from the very charging of the Leyden phial, it seems clear, that the electrical fluid does in reality not pervade the substance of glass, yet it is still difficult to conceive how such a subtile fluid may be forced out from one side of a very thick pane of glass, by a similar quantity of electrical fire thrown upon the other surface, and yet that it does not pass through any substance of glass, however thin, without breaking it. Is there some other fact or illustration besides those to be found in your public writings, by which it may be made more. obvious to our understanding, that electrical fire does not enter at all the very substance of glass, and yet may force from the opposite surface an equal quantity; or that it really enters the pores of the glass without breaking it? Is there any comparative illustration or example in nature, by which it may be made clear, that

a fluid thrown upon one surface of any body, may force out the same fluid from the other surface without passing through the substance?

ANSWER.

That the electric fluid, by its repulsive nature, is capable of forcing portions of the same fluid out of bodies without entering them itself, appears from this experiment. Approach an isolated body with a rubbed tube of glass; the side next the tube will then be electrized negatively, the opposite positively. If a pair of cork balls hang from that opposite side, the electrical fluid forced out of the body will appear in those balls, causing them to diverge. Touch that Touch that opposite side, and you thereby take away the positive electricity. Then remove the tube, and you leave the body all in a negative state. Hence it appears, that the electric fluid appertaining to the glass tube did not enter the body, but retired with the tube, otherwise it would have supplied the body with the electricity it had lost.

With regard to powder magazines, my idea is, that to prevent the mischief which might be occasioned by the stones of their walls flying about in case of accidental explosion, they should be constructed in the ground; that the walls should be lined with lead, the floor lead, all a quarter of an inch thick, and the joints well soldered; the cover copper, with a little scuttle to enter the whole, in the form of a canister for tea. If the edges of the cover-scuttle fall into a copper channel containing mercury, not the smallest particle of air or moisture can enter to the powder, even though the walls stood in water, or the whole was under water.

TO JOHN INGENHOUSZ.

An Attempt to explain the Effects of Lightning on the Vane of the Steeple of a Church in Cremona, August, 1777.

1. WHEN the subtile fluid, which we call fire or heat, enters a solid body, it separates the particles of which that body consists farther from each other, and thus dilates the body, increasing its dimensions.

2. A greater proportion of fire introduced separates the parts so far from each other, that the solid body becomes a fluid, being melted.

3. A still greater quantity of heat separates the parts so far, that they lose their mutual attraction, and acquire a mutual repulsion, whence they fly from each other, either gradually or suddenly, with great force, as the separating power is introduced gradually or suddenly.

4. Thus ice becomes water, and water vapor, which vapor is said to expand fourteen thousand times the space it occupied in the form of water, and with an explosive force in certain cases capable of producing great and violent effects.

5. Thus metals expand, melt, and explode; the two first effected by the gradual application of the separating power, and all three, in its sudden application, by artificial electricity or lightning.

6. That fluid in passing through a metal rod or wire is generally supposed to occupy the whole dimension of the rod. If the rod is smaller in some places than in others, the quantity of fluid, which is not sufficient to make any change in the larger or thicker part, may be sufficient to expand, melt, or explode the smaller, the quantity of fluid passing being the same, and the quan tity of matter less that is acted upon.