Benjamin Franklin Papers

Experiments Supporting the Use of Pointed Lightning Rods, [18?] August 1772

Experiments Supporting the Use of Pointed Lightning Rods

Printed in Benjamin Vaughan, ed., Political, Miscellaneous, and Philosophical Pieces … by Benj. Franklin … (London, 1779), pp. 487–99; French translation in Jacques Barbeu-Dubourg, Œuvres de M. Franklin … (2 vols., Paris, 1773), I, 289–300; ADS (lacking the conclusion of the published version): Bibliothèque de Genève; AD (fragment of the published conclusion): American Philosophical Society

The following document raises its full share of editorial problems, but none of them greatly affects its substance. The bulk of the paper was unquestionably what Franklin read to the Purfleet committee, and what brought the majority of its members to accept his position.2 When the experiments were first published, in Dubourg’s French translation, the editor said that they were read before the committee and dated them August 27; six years later Vaughan published the English text with a note, apparently in the copy from which he was working, that said the same thing.3 Neither editor knew, however, that Franklin had sent another version to his Swiss friend, de Saussure, in the autumn of 1772. It was, according to him, what he had read to the committee; and he dated it August 18. It shortens the description of the sixth experiment in the printed text and omits that text’s concluding paragraph.4 Both documents, in other words, have Franklin’s imprimatur, one indirectly through editions that he supervised and the other in his own hand; yet the two differ in date and in substance. When did he actually read his paper to the committee, and in what form?

On August 12, the earliest meeting of which we have record, Benjamin Wilson objected to the use of pointed lightning rods at Purfleet. On the 17th the committee considered the objection and on the following day, in the light of Franklin’s arguments, rejected it and asked the American to complete the draft report. On the 21st the committee met to sign the report, Wilson dissenting;5 we have no evidence whatever that it met again until December. Hence August 27, despite Franklin’s imprimatur, is an impossible date for the paper. It was presumably written between the 12th, when Wilson first objected, and the 18th, when the issue was settled.

Which form Franklin used to convince the committee is impossible to say, but we are inclined to believe that it was the longer one printed here. Dubourg and Vaughan said so. In its conclusion, furthermore, which chiefly distinguishes it from the shorter version, Franklin rebuts one of Wilson’s main arguments, and in polite terms describes it as nonsense; before the committee, it seems to us, he would have been likely to attack all his opponent’s points. For these reasons we accept the printed text, but for its date we substitute that of the shortened and slightly altered version that Franklin sent to Switzerland. Why did he shorten it? Perhaps because the full argument seemed unnecessary: de Saussure was so expert in electrical matters, Franklin may have believed, that for him a conclusion arguing against nonsense would have belabored the obvious.

Aug. 27 [i.e., 18?], 1772.

Experiments, Observations, and Facts, tending to support the opinion of the utility of long pointed rods, for securing buildings from damage by strokes of lightning.*

Experiment I.

The prime conductor of an electric machine, A. B. being supported about 10½ inches above the table by a wax-stand,6 and under it erected a pointed wire 7½ inches high and ⅕ of an inch7 thick, tapering to a sharp point, and communicating with the table; When the point (being uppermost) is covered by the end of a finger, the conductor may be full charged, and the electrometer c, will rise to the height indicating a full charge: But the moment the point is uncovered, the ball of the electrometer drops, shewing the prime conductor to be instantly discharged and nearly emptied of its electricity. Turn the wire its blunt end upwards, (which represents an unpointed bar,) and no such effect follows, the electrometer remaining at its usual height when the prime conductor is charged.


What quantity of lightning, a high pointed rod well communicating with the earth may be expected to discharge from the clouds silently in a short time, is yet unknown; but I have reason from a particular fact to think it may at some times be very great. In Philadelphia I had such a rod fixed to the top of my chimney, and extending about nine feet above it. From the foot of this rod, a wire (the thickness of a goose quill) came through a covered glass tube in the roof, and down through the well of the stair-case; the lower end connected with the iron spear of a pump. On the staircase opposite to my chamber-door, the wire was divided; the ends separated about six inches, a little bell on each end; [and]9 between the bells a little brass ball suspended by a silk thread, to play between and strike the bells when clouds passed with electricity in them. After having frequently drawn sparks and charged bottles from the bell of the upper wire, I was one night waked by loud cracks on the stair-case. Starting up and opening the door, I perceived that the brass ball, instead of vibrating as usual between the bells, was repelled and kept at a distance from both; while the fire passed sometimes in very large quick cracks from bell to bell; and sometimes in a continued dense white stream, seemingly as large as my finger, whereby the whole stair-case was enlightened as with sunshine, so that one might see to pick up a pin.* 1 And from the apparent quantity thus discharged, I cannot but conceive that a number* of such conductors must considerably lessen that of any approaching cloud, before it comes so near as to deliver its contents in a general stroke: An effect not to be expected from bars unpointed; if the above experiment with the blunt end of the wire is deemed pertinent to the case.

Experiment II.

The pointed wire under the prime conductor continuing of the same height, pinch it between the thumb and finger near the top, so as just to conceal the point; then turning the globe, the electrometer will rise and mark the full charge. Slip the fingers down so as to discover about half an inch of the wire, then another half inch, and then another; at every one of these motions discovering more and more of the pointed wire; you will see the electrometer fall quick and proportionably, stopping when you stop. If you slip down the whole distance at once, the ball falls instantly down to the stem.


From this experiment it seems that a greater effect in drawing off the lightning from the clouds may be expected from long pointed rods, than from short ones; I mean from such as show the greatest length, above the building they are fixed on.

Experiment III.

Instead of pinching the point between the thumb and finger, as in the last experiment, keep the thumb and finger each at near an inch distance from it, but at the same height, the point between them. In this situation, though the point is fairly exposed to the prime conductor, it has little or no effect; the electrometer rises to the height of a full charge. But the moment the fingers are taken away, the ball falls quick to the stem.


To explain this, it is supposed, that one reason of the sudden effect produced by a long naked pointed wire is, that ( by the repulsive power of the positive charge in the prime conductor) the natural quantity of electricity contained in the pointed wire is driven down into the earth, and the point of the wire made strongly negative; whence it attracts the electricity of the prime conductor more strongly than bodies in their natural state would do; the small quantity of common matter in the point, not being able by its attractive force to retain its natural quantity of the electric fluid, against the force of that repulsion. But the finger and thumb being substantial and blunt bodies, though as near the prime conductor, hold up better their own natural quantity against the force of that repulsion; and so, continuing nearly in the natural state, they jointly operate on the electric fluid in the point, opposing its descent, and aiding the point to retain it;4 contrary to the repelling power of the prime conductor, which would drive it down. And this may also serve to explain the different powers of the point in the preceding experiment,5 on the slipping down the finger and thumb to different distances.

Hence is collected, that a pointed rod erected between two tall chimnies, and very little higher, (an instance of which I have seen) cannot have so good an effect, as if it had been erected on one of the chimneys, its whole length above it.6

Experiment IV.

If, instead of a long pointed wire, a large solid body, (to represent a building without a point) be brought under and as near the prime conductor, when charged; the ball of the electrometer will fall a little; and on taking away the large body, will rise again.


Its rising again shows that the prime conductor lost little or none of its electric charge, as it had done through the point: The falling of the ball while the large body was under the conductor, therefore shows that a quantity of its atmosphere was drawn from the end where the electrometer is placed7 to the part immediately over the large body, and there accumulated ready to strike into it with its whole undiminished force, as soon as within the striking distance; and, were the prime conductor moveable like a cloud, it would approach the body by attraction till within that distance.8 The swift motion of clouds, as driven by the winds, probably prevents this happening so often as otherwise it might do; for, though parts of the cloud may stoop towards a building as they pass, in consequence of such attraction, yet they are carried forward beyond the striking distance before they could by their descending come within it.

Experiment V.9

Attach a small light lock of cotton to the underside of the prime conductor, so that it may hang down towards the pointed wire mentioned in the first experiment. Cover the point with your finger, and the globe being turned, the cotton will extend itself, stretching down towards the finger as at a; but on uncovering the point, it instantly flies up to the prime conductor, as at b, and continues there as long as the point is uncovered. The moment you cover it again, the cotton flies down again, extending itself towards the finger; and the same happens in degree, if (instead of the finger) you use, uncovered, the blunt end of the wire uppermost.


To explain this, it is supposed that the cotton, by its connection with the prime conductor, receives from it a quantity of its electricity; which occasions its being attracted by the finger that remains still in nearly its natural state. But when a point is opposed to the cotton, its electricity is thereby taken from it, faster than it can at a distance be supplied with a fresh quantity from the conductor. Therefore being reduced nearer to the natural state, it is attracted up to the electrified prime conductor; rather than down, as before, to the finger.

Supposing farther that the prime conductor represents a cloud charged with the electric fluid; the cotton, a ragged fragment of cloud (of which the underside of great thunder clouds are seen to have many;) the finger, a chimney or highest part of a building. We then may conceive that when such a cloud passes over a building, some one of its ragged under-hanging fragments may be drawn down by the chimney or other high part of the edifice; creating thereby a more easy communication between it and the great cloud. But a long pointed rod being presented to this fragment, may occasion its receding, like the cotton, up to the great cloud; and thereby increase, instead of lessening the distance, so as often to make it greater than the striking distance. Turning the blunt end of a wire uppermost, (which represents the unpointed bar) it appears that the same good effect is not from that to be expected. A long pointed rod it is therefore imagined, may prevent some strokes; as well as conduct others that fall upon it, when a great body of cloud comes on so heavily that the above repelling operation on fragments cannot take place.

Experiment VI.1

Opposite the side of the prime conductor place separately, isolated by wax stems, Mr. Canton’s two boxes with pith balls suspended by fine linen threads. On each box, lay a wire six inches long and ⅕ of an inch thick, tapering to a sharp point; but so laid, as that four inches of the pointed end of one wire, and an equal length of the blunt end of the other, may project beyond the ends of the boxes; and both at 18 inches distance from the prime conductor. Then charging the prime conductor by a turn or two of the globe, the balls of each pair will separate; those of the box whence the point projects most, considerably; the others less. Touch the prime conductor, and those of the box with the blunt point will collapse, and join. Those connected with the point will at the same time approach each other, till within about an inch, and there remain.


This seems a proof, that though the small sharpened part of the wire must have had a less natural quantity in it before the operation, than the thick blunt part; yet a greater quantity was driven down from it to the balls. Thence it is again inferred that the pointed rod is rendered more negative: and farther, that if a stroke must fall from the cloud over a building, furnished with such a rod, it is more likely to be drawn to that pointed rod, than to a blunt one; as being more strongly negative, and of course its attraction stronger. And it seems more eligible, that the lightning should fall on the point of the conductor (provided to convey it into the earth,) than on any other part of the building, thence to proceed to such conductor. Which end is also more likely to be obtained by the length and loftiness of the rod; as protecting more extensively the building under it.

It has been objected,2 that erecting pointed rods upon edifices, is to invite and draw the lightning into them; and therefore dangerous. Were such rods to be erected on buildings, without continuing the communication quite down into the moist earth, this objection might then have weight; but when such compleat conductors are made, the lightning is invited not into the building, but into the earth, the situation it aims at; and which it always seizes every help to obtain, even from broken partial metalline conductors.

It has also been suggested, that from such electric experiments nothing certain can be concluded as to the great operations of nature; since it is often seen that experiments, which have succeeded in small, in large have failed. It is true that in mechanics this has sometimes happened. But when it is considered that we owe our first knowledge of the nature and operations of lightning, to observations on such small experiments; and that on carefully comparing the most accurate accounts of former facts, and the exactest relations of those that have occurred since, the effects have surprizingly agreed with the theory; it is humbly conceived that in natural philosophy, in this branch of it at least, the suggestion has not so much weight; and that the farther new experiments now adduced in recommendation of long sharp-pointed rods, may have some claim to credit and consideration.

It has been urged too, that though points may have considerable effects on a small prime conductor at small distances; yet on great clouds and at great distances, nothing is to be expected from them. To this it is answered, that in those small experiments it is evident the points act at a greater than the striking distance; and in the large way, their service is only expected where there is such nearness of the cloud, as to endanger a stroke; and there, it cannot be doubted the points must have some effect.3 And if the quantity discharged by a single pointed rod may be so considerable as I have shown it; the quantity discharged by a number, will be proportionably greater.

But this part of the theory does not depend alone on small experiments. Since the practice of erecting pointed rods in America, (now near 20 years) five of them have been struck by lightning; viz. Mr. Raven’s and Mr. Maine’s in South Carolina; Mr. Tucker’s in Virginia; Mr. West’s and Mr. Moulder’s in Philadelphia.4 Possibly there may have been more that have not come to my knowledge. But in every one of these, the lightning did not fall upon the body of the house, but precisely on the several points of the rods; and, though the conductors were sometimes not sufficiently large and compleat, was conveyed into the earth, without any material damage to the buildings. Facts then in great, as far as we have them authenticated, justify the opinion that is drawn from the experiments in small as above related.5

It has also been objected, that unless we knew the quantity that might possibly be discharged at one stroke from the clouds, we cannot be sure we have provided sufficient conductors; and therefore cannot depend on their conveying away all that may fall on their points.6 Indeed we have nothing to form a judgment by in this case but past facts; and we know of no instance where a compleat conductor to the moist earth has been insufficient, if half an inch diameter. It is probable that many strokes of lightning have been conveyed through the common leaden pipes affixed to houses to carry down the water from the roof to the ground: and there is no account of such pipes being melted and destroyed, as must sometimes have happened if they had been insufficient. We can then only judge of the dimensions proper for a conductor of lightning, as we do of those proper for a conductor of rain, by past observation. And as we think a pipe of three inches bore sufficient to carry off the rain that falls on a square of 20 feet, because we never saw such a pipe glutted by any shower; so we may judge a conductor of an inch diameter, more than sufficient for any stroke of lightning that will fall on its point. It is true that if another deluge should happen wherein the windows of heaven are to be opened, such pipes may be unequal to the falling quantity; and if God for our sins should think fit to rain fire upon us, as upon some cities of old, it is not expected that our conductors of whatever size, should secure our houses against a miracle. Probably as water drawn up into the air and there forming clouds, is disposed to fall again in rain by its natural gravity, as soon as a number of particles sufficient to make a drop can get together;7 so when the clouds are (by whatever means) over or undercharged [with the electric fluid] to a degree sufficient to attract them towards the earth, the equilibrium is restored, before the difference becomes great beyond that degree.8 Mr. Lane’s electrometer,1 for limiting precisely the quantity of a shock that is to be administered in a medical view, may serve to make this more easily intelligible. The discharging knob does by a screw approach the conductor to the distance intended, but there remains fixed. Whatever power there may be in the glass globe to collect the fulminating fluid, and whatever capacity of receiving and accumulating it there may be in the bottle or glass jar; yet neither the accumulation or the discharge, ever exceeds the destined quantity. Thus, were the clouds always at a certain fixed distance from the earth, all discharges would be made when the quantity accumulated was equal to the distance: But there is a circumstance which by occasionally lessening the distance, lessens the discharge; to wit, the moveableness of the clouds, and their being drawn nearer to the earth by attraction when electrified; so that discharges are thereby rendered more frequent and of course less violent. Hence whatever the quantity may be in nature, and whatever the power in the clouds of collecting it; yet an accumulation and force beyond what mankind has hitherto been acquainted with, is scarce to be expected.

B. F.

[Note numbering follows the Franklin Papers source.]

2See the headnotes on BF to Dawson above, May 29, and on the committee’s report below, Aug. 21.

3The note is the first of the two below. Vaughan was scrupulous about enclosing his editorial comments in brackets; their absence convinces us that this note was on the MS. Dubourg’s comment is that Wilson “proposa les raisons sur lesquelles il se fondoit, et M. Franklin y répondit de la maniere que l’on verra ci-après, qui eut l’approbation des trois autres Commissaires. …” Œuvres, I, 288. Vaughan always worked from a MS source unless otherwise indicated, and he gave no such indication in printing these experiments. If, then, he was not retranslating Dubourg, yet produced a text virtually identical with his, the two editors must have been using copies of a single MS. Dubourg’s copy was presumably dated Aug. 27, and Vaughan’s certainly was. We print the English text (without its over abundant italics) as being closer to this original, now lost, than the French translation.

4BF’s letter to de Saussure below, Oct. 8, enclosed this ADS. Both documents have recently been published, with an introduction, by Paul A. Tunbridge, “Franklin’s Pointed Lightning Conductor,” Royal Soc., Notes and Records, XXVIII (1974), 207–19. Tunbridge believes that the ADS was an earlier version of BF’s final paper, but adduces no evidence to dissuade us from our guess that it was later.

5See the committee’s minutes above, under Aug. 12, and its report below, Aug. 21.

8Notes printed in this manner are not in the ADS (which has only one note, indicated below) and are not, we believe, by Vaughan. Relevant portions of his own notes we identify and incorporate in our annotation.

6The ADS, which except for the date is virtually identical as far as it goes with the printed text, here reads “Glass Stand.”

7The ADS reads “¼.”

9The bracketed word is not in the ADS, and a bracketed phrase in the final paragraph is not in the AD; these are presumably Vaughan’s insertions.

2For Jacques de Romas, the experimental physicist, see above, V, 396 n.

1BF devised his apparatus in 1752, and some years later John Winthrop copied it: above, V, 69–71; X, 150. The ball oscillated when the two bells were oppositely charged (indicating a negative charge in the cloud) and rang them in a way that disturbed DF: above, VIII, 94. When the charges were the same (indicating a positively charged cloud) the current flowed across the gap; DF’s attitude toward the resultant illumination can be imagined.

3BF appended a different note in the ADS: “The Magazines in question are five distinct Buildings each 150 feet long, and 50 feet distant from each other. It was proposed to have a pointed Conductor erected at each End of each Building; in all 10 Conductors, connected by Lead along the Ridge of the Roofs.” BF made this proposal after his first visit to Purfleet; see his letter to Dawson above, May 29.

4Vaughan here commented: “Perhaps their first and principal tendency is, to repel and thereby lessen the influence of the fluid in the conductor.”

5For BF’s doctrine of points see above, III, 127–8, 472–3; V, 14–20, 23–4.

6BF had made this point before: above, X, 56; XIV, 263. Wilson disagreed; see the headnote on the committee’s report below, Aug. 21.

7“I.e.,” Vaughan elaborated, “drawn for a time, to a different part of the conductor, but not out of it.”

8For Henly’s supporting experiments see the headnote on the committee’s report. BF defined striking distance (above, XIV, 262) as the maximum distance across which a discharge will take place from one body to another.

9BF had performed this experiment more than twenty years before: above, V, 78; see also X, 56.

1This is the one point where Vaughan’s text differs markedly from the ADS, but in detail rather than substance. “Supporting the pointed Wire with its Stand,” the ADS reads, “on Wax, contrive to suspend a Pair of the larger Pith Balls hanging so freely as that they may separate widely on occasion, when the pointed Wire is under the Prime Conductor. Turn the Globe while the Point is uppermost, observing the Distance to which the Balls recede from each other. Then turn up the blunt End of the Wire, and again observe the Distance. It will be found small in the latter Case.” The experiment was not BF’s but Henly’s, and had been performed the previous April as part of a series; see the headnote on the committee’s report below, Aug. 21.

2The objections that follow were Wilson’s, made in the committee and outside it; his arguments are discussed in ibid.

3Because they will either draw off the electricity and prevent a stroke, or conduct it safely. The “small experiments” were Henly’s the previous spring.

4Tucker we cannot identify; for Raven, Maine, West, and Moulder see above, respectively, X, 53 n, 54 n; IX, 291 n; XVII, 249 n.

5The ADS ends here, with the date and BF’s signature. On the verso is another sketch by him of Henly’s electrometer; the only information in it that is not in the sketch above (XVIII, 183) is that the arm may be made of boxwood instead of ivory. BF’s fragmentary AD begins with the paragraph following this one.

6Wilson had raised this objection years before and was still harping on it; see the headnote on the committee’s report below, Aug. 21.

7See above, XVIII, 154–7.

8BF’s fragmentary AD ends here.

1Fully described above by its inventor: XIII, 460–2.

Authorial notes

[The following note(s) appeared in the margins or otherwise outside the text flow in the original source, and have been moved here for purposes of the digital edition.]

º *Read at the committee appointed to consider the erecting conductors to secure the magazines at Purfleet. Aug. 27, 1772.8

º †Mr. Henley’s.

º *Mr. De Romas2 saw still greater quantities of lightning brought down by the wire of his kite. He had “explosions from it, the noise of which greatly resembled that of thunder, and were heard (from without) into the heart of the city, notwithstanding the various noises there. The fire seen at the instant of the explosion had the shape of a spindle eight inches long and five lines in diameter. Yet from the time of the explosion to the end of the experiment, no lightning was seen above, nor any thunder heard. At another time the streams of fire issuing from it were observed to be an inch thick and ten feet long.” See Dr. Priestley’s History of Electricity, pages 354–6. first edition.

º *Twelve were proposed on or near the magazines at Purfleet.3

Index Entries