and promising myself if otherwise, the honor at least of acknowledging frankly my error, and of being thankful to him who kindly shows it to me.

16. By the account given of this stroke of lightning upon the steeple at Cremona, it appears that the rod of iron or spindle on which the vane turned was of about two inches circumference, terminating in a cross above the vane, and its lower end fixed in a marble pedestal.

17. That the plate of the vane was copper, eight or nine inches wide, and near twice as long. That it was about one line thick near the spindle and growing thinner insensibly towards the other end, where its thickness did not exceed three quarters of a line, the weight 20 ounces.

18. That the copper had been tinned over.

19. That the marble pedestal was split by the stroke into many pieces, and scattered over the roof, garden, and court of a neighboring building. One piece was thrown to the distance of 40 feet. The spindle was broken and displaced, and the vane thrown on the roof of the parsonage house 20 feet from the steeple.

20. That the vane was perforated in 18 places, the holes of irregular forms, and the metal which had filled them pushed outwards in some of them on one side of the vane, in others on the other. The copper shewed marks of having been partly melted, and in some places tin and copper melted and mixed together. There were marks of smoke in several places.

21. The ragged parts bent outwards round each hole, being brought back to their original flat position, were not, though evidently a little thinned and dilated, sufficient to fill the place.

22. From the effects described (19,) it is clear that the quantity of lightning which fell on this steeple at Cremona was very great.

23. The vane being a thin plate of copper, its edges and corners may be considered as a series of points, and being therefore sooner rendered negative by the repulsive force of an approaching positive cloud than the blunt and thick iron cross, (12,) was probably first struck; and thence became the conductor of that great quantity.

24. The plate of which the vane was formed being thicker near the spindle, and diminishing in thickness gradually to the other end, (17,) was probably not of copper plated by passing between rollers, for they would have left it of equal thickness; but of metal plated by the hammer. The surface too of rolled

copper is even and plain, that of hammered is generally uneven, with hollows occasioned by the impressions of the hammer.

25. In those concave impressions the metal is thinner than it is around them, and probably thinnest near the center of each impression.

26. The lightning which in passing through the vane was not sufficient to melt its thicker parts, might be sufficient to melt the thinner, (6), (7), (8), (9), and to soften those that were in a middle state.

27. The part of the tin (18) which covered the thinner parts being more easily melted and exploded than copper, (10) might possibly be exploded when the copper was but melted. The smoke appearing in several placès, (20) is a proof of explosion.

28. There might probably be more tin in the concave impressions of the hammer on one side of the plate, than on the convex part of those impressions on the other. Hence stronger explosions on the concave side.

29. The nature of those explosions is to act violently in all directions; and in this case being near the plate they would act against it on one side, while they acted against the air on the other.

30. These thin parts of the plate being at the same instant partly in fusion, and partly so softened as to be near it; the softened parts were pushed outwards, a hole made, and some of the melted parts blown away; hence there was not left metal enough to refill the vacancy by bending back the ragged parts to their places.

31. The concave impressions of the hammer being indifferently made on both sides of the plate, it is natural from 28, 29, 30, that the pushing outwards of the softened metal by explosions, should be on both sides of the plate nearly equal.

32. That the force of a simple electrical explosion is very great, appears from the Geneva experiment, wherein a spark between two wires, under oil in a drinking glass, breaks the glass, body, stem, and foot, all to shivers.

33. The electric explosion of metal acts with still more force. A strip of leaf gold no broader than a straw, exploded between two pieces of thick looking glass, will break the glass to pieces, though confined by the screws of a strong press. And between two pieces of marble pressed together by a weight of 20 pounds will lift that weight. Much less force is necessary to move the melted and softened parts of a thin plate of copper.

34. This explication of the appearances on the vane, is drawn from what we already know of electricity and the effects of lightning. The learned author of the account gives a different but very ingenious one, which he draws from the appearances themselves. The matter pushed out of the holes is found, that of some on one side of the plate, and of others on the other. Hence he supposes them to be occasioned (if I understand him right) by streams or threads of electric matter of different and contrary kinds, rushing violently towards each other, and meeting with the vane, so accidently placed, as to be found precisely in the place of their meeting, where it was pierced by all of them, they all striking on both its sides at the same instant. This however is so extraordinary an accident, as to be in the authors own opinion almost miraculous, "Passeranno (says he) forse piu secoti prima que ritorni tralle infinite combinazioni un caso similé a quello della banderuola che ora abbiamo per mano. Forza é que si essaurisca una non piu udita miniera difulmini sopra una grande citta, pressoque seminata di campanili e di banderuole, il che e raris simo; e può ancora volti cio succedere, senza che s'incontri giammai un altera, banderuola tanto opportunetamente situata tra i limiti della fulminea explosione."

35. But though the author's explication of these appearances of the vane does not satisfy me, I am not so confident of my own as to propose its being accepted without confirmation by experiment. Those who have strong electric batteries may try it thus: form a little vane of paper, and spot it on both sides by attaching small pieces of leaf gold or tin foil, not exactly opposite to each other: then send the whole force of the battery through the vane, entering at one end of it and going out at the other. If the metal explodes, I imagine it will be found to make holes in the paper, forcing the torn parts out on the sides opposite to the metal. A more expensive but perhaps a more satisfactory experiment would be, to make a new vane as exactly as possible like that in question, in all the particulars of its description, and place it on a tall mast fixed on some hill subject to strokes of lightning, with a better conductor to the earth than the wood of the mast; if this should be struck in the course of a few years, and the same effects appear upon it, it would be still more miraculous to suppose it happened by accident to be exactly situated where those crossing threads of different electricities were afterwards to meet.

36. The perforation of glass bottles when overcharged is, I imagine, a different case, and not explicable by either of these hypothesis. I cannot well suppose the breach to be occasioned by the passage of electricity through it, since a single

hottle, though so broken in the discharge, always is found to send round in its usual course the quantity with which it was charged. Then the breach never happens but at the instant of the circuitous discharge, either by the discharging rod, or in overleaping the borders of the glass. Thus I have been present when a battery of twenty glasses was discharged by the discharging rod, and produced the same effect in its circuit as if none of the bottles had been pierced ; and yet on examining them, we found no less than twelve of them in that situation. Now all the bottles of the battery being united by a communication of all the outsides together, and of all the insides together, if one of them had been pierced by a forced passage of the different kinds of electricity to meet each other, before the discharge by the discharging rod, it would not only have prevented the passage of the electricity by the common circuit, but it would have saved all the rest of its fellows, by conducting the whole through its own breach. And it is not easy to conceive that 12 bottles in 20 should be so equally strong as to support the whole strength of their charge, till the circuit of their discharge was opened, and then be so equally weak as to break altogether when the weight of that charge was taken off from them by opening the circuits. At some other time I will give you my opinion of this effect if you desire it.

I have taken the account of this stroke of lightning from an Italian piece, intitled Analisi d'un nuovo fenomeno del fulmine, the dedication of which is subscribed Carlo Barletti delle Sacolo Pic, who I suppose is the author. As I do not perfectly understand that language, I may possibly in some things have mistaken that philosopher's meaning. I therefore desire, my dear friend, that you would not permit this to be published, 'till you have compared and considered it with that original piece, and communicated to me your remarks and corrections. Nor would I in any case have it appear with my name, as perhaps it may occasion disputes, and I have no time to attend to them.

1

ON ELECTRICITY.

THE LEYDEN BOTTLE, AND M. VOLTA'S EXPERIMENT.

To *

Paris, 1778.

I thank you for the account you give me of M. Volta's experiment. You judge rightly in supposing that I have not much time at present to consider

philosophical matters; but as far as I understand it from your description, it is only another form of the Leyden Phial, and explicable by the same principles. I must however own myself puzzled by one part of your account, viz. " and thus the electric force once excited may be kept alive years together," which is perhaps only a mistake. I have known it indeed to be continued many months in a phial hermetically sealed, and suppose it may be so preserved for ages; but though one may, by repeatedly touching the knob of a charged bottle with a small insulated plate, like the upper one of the electrophore, draw an incredible number of sparks successively, that is, one after every touch, and those for a while not apparently different in magnitude, yet at length they will become small, and the charge be finally exhausted. But I am in the wrong to give my opinion till I have seen the experiment.

I like much your pasteboard machine, and think it may, in some respects, be preferable to the very large glass ones constructed here. The Duke de Chaulnes has one, said, if I remember right, to be five feet in diameter. I saw it tried, but it happened not to be in order.

AURORA BOREALIS.

B. F.

Suppositions and Conjectures towards forming an Hypothesis for its Explanation,

1779.

1. Air heated by any means becomes rarefied and specifically lighter than other air in the same situation not heated.

2. Air being thus made lighter rises, and the neighboring cooler heavier air takes its place.

3. If in the middle of a room you heat the air by a stove, or pot of burning coals near the floor, the heated air will rise to the ceiling, spread there over the cooler air till it comes to the cold walls; there being condensed and made heavier, it descends to supply the place of that cool air which had moved towards the stove or fire, in order to supply the place of the heated air which had ascended from the space around the stove or fire.

4. Thus there will be a continual circulation of air in the room, which may be rendered visible by making a little smoke; for that smoke will rise and circulate with the air.

5. A similar operation is performed by nature on the air of the globe. Our atmosphere is of a certain height, perhaps at a medium [ ] miles. Above