Lodestone, a ferruginous rock similar in weight and color to the kind of iron ore called in the rock . [1] It contains iron in more or less considerable quantity, and it is in this metal mixed with salt and oil that the magnetic property [ vertu ] resides rather than in its rocky substance. [2] This famous rock was known to the Ancients, for we know from the testimony of Aristotle that Thales, the most ancient Greek philosopher, spoke of the lodestone; but it is not certain that the name used by Aristotle is the one used by Thales. It is Onomacritus, who lived in the LX ^th olympiad, [3] of whom we have a few poems under the name of Orpheus, who furnishes us with the most ancient name of the lodestone: he calls it μαγνήτης. Hippocrates ( Lib. de sterilib. mulier. ) designated the lodestone under the periphrase stone that attracts iron : λίθος ἥτις τὸν σίδηρον ἁρπάζει.

The Arabs and the Portuguese use the same periphrase, which Sextus Empiricus expressed in a single word: σιδηραγωγός. Sophocles, in one of his plays that has not come down to us, had named the lodestone Lydia stone . Hesychius preserved this word for us as well as, Λυδικὴ λίθος, which is a variation. Plato, in the Timaeus , calls the lodestone stone of Heracles , one of the names most used among the Greeks.

Aristotle did the lodestone more honor than anyone by giving it no name; he calls it ἡ λίθος, the stone par excellence . Themipius expresses himself in the same way. Theophrastus, with most of the Ancients, followed the already established appellation λίθος Ήρακλεία.

Pliny, because of a misunderstood passage of Theophrastus, thought that the touchstone, coticula , which has among other names that of Λυδὴ λίθος, in addition had the name Ήρακλεία in common with the lodestone. The Greeks and the Latins also made use of the word σιδηρ τις derived from σίδηρος iron , whence the old French name pierre ferriere . Finally, the Greeks diversified the name of μαγνήτης in various ways: in Taetzès we find μάγνησσα λίθος, in Achilles Tatius μαγνησία; μαγν τις in most authors, μαγν τις in a few, as well as ὁ λίθος μαγνίτης, by the permutation of η to ι, familiar to the Greeks from the earliest times; and μάγνης, which is not of all these names the one they most used, is almost the only one that passed on to the Latins.

As for the origin of this denomination of the lodestone, it comes manifestly from the place where the lodestone was first discovered. There were two cities in Asia Minor called Magnetie , one near Meander, the other under Mount Sipylus. The latter, which belonged particularly to Lydia, and which was also called Heraclea , according to the testimony of Aelius Dionysius in Eustathius, was the true homeland of the lodestone. Mount Sipylus was no doubt rich in metals, and consequently in lodestone; thus the lodestone called magnes for the place of its first discovery preserved its ancient name, as has occurred with steel and copper, which bear the names of the places where they were discovered. What is singular is that the worst of the five lodestone varieties reported by Pliny was the one from Asia Minor, the first homeland of the lodestone, as the best of all was from Ethiopia.

Marbodaeus says that the lodestone was found among the Troglodytes, and that this stone also comes from the Indies. Isidore of Seville says that the Indians knew it first, and after him most of the writers of the Middle and Late Middle Ages called the lodestone lapis Indicus , giving the homeland of the species to the whole genus.

The Ancients knew really nothing about the lodestone except its property of attracting iron; that was the principal subject of their admiration, as we can see from this passage in Pliny: Quid lapidis rigore pigrius? Ecce sensus manusque tribuit illi natura. Quid ferri duritie pugnacius? Sed cedit et patitur mores: Trahitur namque à magnete lapide, domitrixque illa rerum omnium materia ad inane nescio quid currit, atque ut propius venit, assistit teneturque, et complexu haeret. (Pliny, book XXXVI, chap. xvi). [4]

However, it appears that they knew something of its communicative property. Plato gives an example in the Ion , where he describes that famous chain of iron rings suspended from each other, the first of them holding to the lodestone. Lucretius, Philo, Pliny, Galien, and Nemesius, report the same phenomenon, and Lucretius also mentions the propagation of the magnetic property through the hardest bodies, as appears in these verses:

But we do not see in any passage of their writings that they knew anything of the directive property of the lodestone;  [6] we know absolutely nothing about when this discovery was made, and we do not even know exactly when it was applied to the uses of navigation.

There is every reason to believe that chance led someone to the discovery that the lodestone placed on water in a small boat constantly oriented itself north and south, and that a piece of magnetized iron had the same property; that this magnetized iron was placed on a pivot so it could move more freely; that then it was thought this discovery could well be useful to navigators to ascertain south and north when the weather was overcast and no star could be seen; finally that the ordinary compass was substituted for the magnetized needle to remedy the displacements occasioned by the shaking of the ship. It also appears that this discovery was made before the year 1180. See the article Needle, where this discovery is more particularly treated.

I. On the poles of the lodestone and its directive property.

Every lodestone has two poles where most of its power resides. They are recognized by rolling any lodestone in some iron filings: all the parts of these filings that attach to the stone orient themselves toward one or the other of these poles, and those that are immediately on them at these points stand up perpendicularly on the stone; in short, the filings are more strongly attracted, and in greater quantity, to the poles than anywhere else. Here is another way to identify the poles: you place a lodestone on a piece of polished glass under which you have placed a white paper; you scatter filings slowly over this glass around the lodestone, and tap the edges of the glass gently to diminish the friction that would prevent the molecules of filings from obeying the magnetic currents; at once you see the filings assume a regular arrangement such as we observe in the figure, in which the filings align in curved lines A E B, A E B (Physics plates, fig. 58 ), as they are farther from the poles, and in straight lines A A, B B , the closer they are, so that the poles are the points where all these different curved and straight lines converge.

Now we call the straight line that runs through the lodestone from pole to pole its axis , and the lodestone’s equator is the perpendicular plane that divides it in the middle of its axis. This property of the lodestone of having poles seems essential to all lodestones; for you could break a lodestone into as many pieces as you might wish, and there will still be two poles in every piece. This polarity of the lodestone does not come, as was believed, from the north-south orientation of the lodestone mines, for it is quite certain that these mines may assume, like others, any sort of orientation; and notably, there is in Devonshire a lodestone mine with its veins oriented from east to west , and the poles also are oriented in that direction. But the lodestone’s poles must not be regarded as two such invariable points that they cannot change place; for Mr. Boyle says that the poles of a small piece of lodestone can be changed by applying them against the more vigorous poles of another stone, which has been confirmed in our time by Mr. Gwarin Knight, who can change the poles of a natural lodestone at will by means of magnetized iron bars. [7]

The poles of the lodestone have been given the same names as the poles of the earth because the freed lodestone has the property of always orienting its poles towards those of our globe; in other words, a lodestone that floats freely on still waters, or which is mobile at its center of gravity, having its axis parallel to the horizon, will constantly come to rest in a situation such that one of its poles always faces north, and the other south; and if this situation is disturbed, even by giving it the opposite orientation, it will not cease to move and oscillate until it has found again its original orientation. The convention is to call austral pole of the lodestone the one that turns northward, and boreal pole the one that turns to the south. The magnetic meridian is the plane perpendicular to the lodestone along the length of its axis, which consequently passes through the poles.

When, after clearly recognizing the poles and the axis of a lodestone, it is allowed to float freely on a cork, the vessel on which it floats being placed on an exactly traced meridian, one will note that the poles of the lodestone do not precisely line up with the earth’s poles, but decline more or less to the east or the west, depending on the different places on earth where this observation is made. This declination of the lodestone also varies by the year, the month, the day, and even the hour in the same place. See the article Needle, where this is treated more particularly.

Similarly, if you place a spherical lodestone in mercury, after having carefully located the axis and poles, it will orient itself at once close to north and south; but you will also note that its axis will incline in a constant manner, such that in our climes the austral pole dips and the boreal pole rises, and the opposite in the other hemisphere. This dip also varies in all places on earth and at all times of the year, as you can see in the article Needle, where it is more amply discussed.

The poles of the lodestone are, as we have said earlier, variable points which we are sometimes free to produce at will and without the aid of any lodestone, as we shall see easily done by the means which we will explain below. For when you cut a lodestone gently and without exertion through the middle of its axis, each of its parts constantly has two poles and becomes a complete lodestone; the parts which were contiguous under the equator before the division, and which were anything but poles, have become poles, and poles of different names, such that each of its parts could equally become the boreal or austral pole, depending on whether the section had taken place closer to the austral or boreal pole of the large lodestone; and the same thing would happen to each of these halves if they were cut in the middle in the same way. See Physics plates fig. 66 .

But if instead of cutting the lodestone through the middle of its axis A B , you cut it longitudinally (Physics plates , fig. 67 ), you will similarly have poles a a, b b , of which those of like name will be in each part on the same side they were on before the division, except that in each part a new axis a b, a b will be formed, parallel to the first, and more or less internalized in the stone, depending on whether it naturally has stronger magnetic strength.

II. On the lodestone’s attractive property

§ I. On the mutual attraction of two lodestones, and on repulsion

The phenomenon of mutual attraction of two lodestones, of one lodestone and a piece of iron, or of two magnetized pieces of iron, is the one which most stimulated the wonder of Ancient philosophers, and led some to say that the lodestone was animate. Indeed, what could be more singular that to see two lodestones seek each other as if by sympathy, approach each other quickly as if eager, unite by one particular side to the point of allowing themselves to be separated only by a considerable force; then in another situation manifest a mutual repulsion that activates them as long as they are together, flee each other as swiftly as they had sought each other, and not be at rest until they are far apart? Such are nevertheless the circumstances of the phenomenon of the lodestone’s attraction and repulsion, as one can easily convince oneself with the following experiment.

Take two lodestones a b , A B (Physics plates , fig 64 ), place each in a small spruce box so they can easily float on still water, protected from air currents; then place them no farther from each other than their sphere of activity [8] extends: you will see them come together with accelerated speed and unite finally in point C , which will be the median of their reciprocal distance if the lodestones are equal in strength and mass, and if the two boxes are perfectly identical. Mark the points b , A by which these lodestones are joined, and separate them by the same distance: they will come together with the same speed and join at the same points. But if you change the situation of one of the lodestones in such a way that it presents to the other the point directly opposite the one which was attracted, they will flee each other with equal speed until they are outside each other’s sphere of activity.

The experiment shows that these two lodestones attract each other by the differently named poles, in other words the boreal pole of one attracts the austral pole of the other, and the latter’s boreal pole the austral pole of the first; on the contrary, the two north poles flee each other as well as the two south poles, so that it is a constant law of magnetism that the mutual and reciprocal attraction takes place by the poles of different names, and repulsion by the poles of the same denomination.

People have tried to determine whether the force that makes these two lodestones approach or flee act on them only up to a specific limit; whether it acts uniformly at all distances within that limit, or, if it was variable, in what proportion it would grow or diminish with relation to different distances. But the result of a large number of experiments has shown that a lodestone’s force extends sometimes farther and sometimes less far. There are some whose activity extends as far as 14 feet, others whose property is undetectable at 8 or 9 inches. The sphere of activity of a given lodestone itself has a variable reach: it is greater on certain days than on others, though it appears that neither the temperature, nor the humidity, nor the dryness of the air play any role in this effect.

Other experiments have shown that near the limits of the sphere of activity, the magnetic force acts at first imperceptibly, that it becomes more considerable as the attracted body approaches the lodestone, and that it is greatest of all at the point of contact; but the proportion of this force at different distances is not the same in different lodestones, which means that we are unable to establish a general rule.

Here is the result of an experiment carefully done by M. du Tour. [9]

He filled a large basin M (Physics plates , fig. 63 ) and with a fork placed on the surface of the water a sewing needle A B which he had magnetized (which we can therefore consider as a lodestone, as we shall see subsequently); he presented a lodestone T at a distance of 13 inches from this needle, which was about the limit of its sphere of activity, and examined the relation of the speeds of the needle at different distances. Here is the result of his observation.

The needle took to cover

the1 ^st inch 120" [10] 7 ^th 28"

2 ^nd 110 8 ^th 16

3 ^rd 70 9 ^th 12

4 ^th 72 10 ^th 6

5 ^th 56 11 ^th 3

6 ^th 44 12 ^th & 13 ^th 1

Total for the 13 inches, 546" = 9' 6".

What has been observed about repulsion is something like the circumstances of the phenomenon of attraction, in other words that the sphere of repulsion varies in different lodestones, as well as the repulsive force at different distances. Several writers have believed that the repulsive force does not extend in any lodestone as far as the attractive force, and is never anywhere as strong as the attractive property, not even at the point of contact, where it is the greatest. The attractive force of the differently named poles of the two lodestones was, going by an observation of M. Musschenbroek, [11] 340 grains at the point of contact,  [12] whereas the repulsive force of the poles of the same name of these two lodestones was only 44 grains at the point of contact of those two poles.

These writers add to these observations another which is no less singular, which is that one finds lodestones (and the same things happens to magnetized bodies) of which the poles of the same name repel each other so much that they are at a median distance from the limits of their sphere of activity, and attract each other on the contrary at the point of contact; others repel each other more forcefully about the middle of their sphere of activity than around the point of contact, where it seems that the repulsion diminishes. Nevertheless, Mr. Mitchell [13] claims to have observed by means of artificial lodestones that the two poles attract and repel each other equally at the same distances, and in every possible direction; that the error of those who believed repulsion weaker than attraction was owing to the fact that one always weakens lodestones and magnetic bodies by approaching them at the poles of the same name, whereas one increases their virtue when they are approached by the poles of different denomination; that this augmentation or diminution of strength occasioned by the proximity of two lodestones becomes imperceptible as they are progressively separated. This is why we see that at a great distance the attraction and repulsion become more and more equal; and reciprocally less equal as the reciprocal distance of the two lodestones diminishes, and they act on each other, in such a way that if a lodestone is strong enough and close enough to damage considerably a weak lodestone that approaches it by the poles of the same name, what occurs is that the pole of the latter will be destroyed and changed to a pole of a different denomination, by means of which the repulsion will be converted into an attraction. Several experiments moreover make it appear to Mr. Mitchell that attraction and repulsion grow and shrink in inverse proportion to the squares of the respective distances of the two poles.

All these effects of reciprocal attraction and repulsion of two lodestones are unimpeded by solid bodies or fluids. The attraction and repulsion of two lodestones was equally strong whether there was a 100-pound mass of lead between them, or nothing but open air. Mr. Boyle has found that the magnetic virtue came through hermetically sealed glass, which is known to be one of the most impenetrable bodies for any sort of particular flow; iron alone seems to intercept the magnetic matter, for a forged iron plate interposed between two lodestones considerably weakens their attractive and repulsive forces.

Similarly, neither wind, nor flame, nor running water interrupts the effects of attraction and repulsion between two lodestones; these actions are as lively in common air as in rarefied or condensed air in a pneumatic machine. Physics plates , fig. 32 and 35 .

§ 2. On the reciprocal attraction of lodestone and iron

The lodestone attracts iron with even more vigor than it attracts another lodestone. If you place on a cork A (Physics plates , fig. 62 ), a piece of cubic iron B which has never been magnetized, and the whole floats on water, and you present to it a lodestone C by either pole, the iron will vigorously approach it; and reciprocally, if you place the lodestone on the cork and present the piece of iron to it, it will come to it equally quickly, so that it appears that the action of the lodestone on the iron, and of the iron on the lodestone, is equal and reciprocal.

This attraction of the lodestone to iron extends to all bodies containing particles of that metal, and their number is very great in nature; it attracts particles of all kinds of soils, sands, stones; salts and sediments of all streams, ashes and soots of all sorts of woods and peats, coals, oils and greases of every variety; honey, wax, beaver, and countless other material. In a word, the lodestone is the touchstone by means of which one detects the smallest ferruginous parts a body contains.

In truth, to discover that these bodies contain iron it is often necessary to use the means of calcination to subject this metal to the lodestone’s action; but this preparation is used only for bodies that do not contain iron in a metallic form, or when its particles are interspersed in a particular manner with other metals: in that case the iron often responds to the action of a very weak lodestone, whereas it resists that of a strong one. Thus was observed in Petersburg an alloy of iron and tin which a weak lodestone attracted, and on which an excellent lodestone had no effect.

No solid or fluid bodies in any way prevent the mutual action of iron and lodestone, if not iron itself, as we have previously noted. Nor does excessive heating of iron diminish these effects, for the boreal pole of a lodestone has been applied to a red-hot roofing nail that was strongly attracted and remained suspended. But it was also observed that great heat in the lodestone lessens its virtue, at least for a time. The lodestone used in the previous experiment was heated, and when it was quite red its boreal pole was applied against another similar roofing nail, which was weakly attracted, although it did remain suspended; nevertheless, after two or three days the stone attracted the nail as powerfully as before it was placed in the fire. The greatest attractive force of a lodestone is around its two poles; there are lodestones that can lift rather considerable nails with their poles, and would be unable to lift the smallest iron filings by their equator. Yet if one turns the different parts of the equator into poles, as we have said takes place by cutting the lodestone into several parts, the attractive force will be very perceptible in these new poles, such that the sum of the weights that a large lodestone can lift once cut into parts will greatly exceed what that piece could lift when it was intact.

§ 3. On lodestone armor [14]

The attractive force of a lodestone fresh out of the mine consists only of having it lift small nails or other pieces of iron of relatively light weight; that is why it is necessary to arm it to increase its strength. Besides, arming combines, orients, and condenses all its power around the poles, and orients all its emanations towards the mass one places under its poles.

It is essential, before arming a lodestone, to locate properly the placement of its poles; for the arming would be useless if it were placed anywhere else than on those parts. Therefore, in order to locate accurately the poles of a lodestone, it will be placed on a smooth white cardboard, and iron filings that are not rusted will be scattered over it, which will be done more evenly by using a sieve; tap gently on the cardboard, and soon you will see a symmetrical arrangement form of the filings that will orient themselves in curved lines E E (Physics plates fig. 58 ) about the equator, following the straight lines A B toward the poles which will be in the two parts of the lodestone where all these straight lines will point. But they can be determined even more precisely by placing on it a very fine and very short needle, for it will stand up perpendicularly at the location of each pole, and will always be oblique at every other point.

Once the lodestone’s poles have been correctly located, it must be sawed in such a way as to be very even and smooth at the location of the poles. Of all the shapes you can give it, the most advantageous will be the one where the axis is the longest, yet without shortening the other dimensions too much.

Now to determine the proportions of the armor, we must first determine the strength of the lodestone we mean to arm; for the stronger it is, the thicker we must make the pieces that constitute its armor. For this effect we will take small, well-polished, and somewhat flat bars of steel which we will apply to one of the poles of the lodestone; to this steel bar we will present immediately below the pole a small ring of iron to which the tray of a balance will be attached, and we will test for the greatest quantity of weight the lodestone can support before the ring to which the plane of the balance is attached lets go of the steel bar; we will do in succession the same experiment with several similar bars, but of different thicknesses, and we will easily find by means of the one that will lift the greatest weight what thickness we need to give to the sheaths of the armor.

When we have determined this thickness, we will pick much thinner and untempered pieces of steel which we will cut in this manner. A B ([Physics plates], fig. 59 ) is one of the legs of this armor, of which the height and width must be equal with respect to the thickness and length of the lodestone; B E D is a sheath of the same piece of steel of which the plane S B D is perpendicular to A B ; its width at the point where it touches the plane A B must be two-thirds of G G , the width of the plate A B , and the thickness of the sheath S E must have the same dimension. Finally, the length B D , which is the quantity by which the sheath will overlap below the stone, will be two-thirds of D S or of S E . This sheath needs to become thinner, and be rounded below from S and D as far as E , so that its width at E will be a third or a fourth of the width S D . It is also very important to pay attention to the thickness of the leg A B : for if it is made too thick or too thin, the armor will have less strength; but that is something that can only be determined by testing, which is why one must proceed as we have done to determine the thickness of the sheath. We observe in general that the top extremity C C must be rounded and a bit less elevated than the lodestone, and that the thickness of the plate must be smaller around C C than around G G . These two plates will then be applied with their sheaths to the respective poles of the lodestone, in such a way that these two pieces touch the lodestone in as many points as possible, and they will be held in place with a tight band of copper and affix to it the suspender X , fig. 60 .

Now to join the attractive force of the two poles, one must have a steel traverse D A C B , very flexible and not tempered, with a length one or two lines greater than the sheaths of the armor,  [15] and about a line thick; there must be a hole with a hook L so that one can suspend the weight which the lodestone can lift.

Once the lodestone has been thus armed, it will be easy to note that its attractive power will be considerably increased; for a given lodestone that could not bear more than a half-ounce in its natural state easily lifts a weight of ten pounds once it is armed; however its emanations do not extend farther when it is armed than when it is in its natural state, as we see from its action on a magnetized needle that can move on its pivot; and if we apply to the feet of the armor the traverse that serves to support the weights which we have the lodestone lift, the distance at which it will act on the needle will be much smaller, the magnetic property turning aside for most part into the traverse.

When we present to an armed lodestone a piece of thick iron rod A B ([Physics plates], fig. 61 ) heavy enough so that the sheath of the armor to which we approach it cannot bear it, and if we take away this piece when the iron rod A B is thus strongly attracted, it will instantly fall, and cease to be supported.

We placed on one of the sheaths of the armor a small plate of polished steel of ten to eleven lines in length, seven in width, and one line in thickness. This plate T ( fig. 61, no. 2 ) bore a small hook on which was suspended the plate of a balance; on the other plate of the armor was placed the traverse G , such that the traverse and the plate touched. Next we put weights in plate S until the lodestone had ceased to support plate T , and found that it took eighteen ounces. Having then removed the traverse and left the plate alone applied against the lodestone, a weight of two ounces in the balance sufficed to separate the plate: which proves that the proximity of the traverse increased by sixteen ounces the attractive power of the pole to which the plate was applied.

Although the attraction of an armed lodestone appears considerable, it happens nevertheless that rather weak causes destroy their effect in an instant. For example, when an oblong piece of iron F ([Physics plates], fig. 68 ) is held under the pole of an excellent lodestone M , and to the lower extremity of this piece of iron is presented the opposite pole of another, weaker lodestone N , the latter will take the iron away from the stronger. One will better judge the outcome of this experiment if it is done on a polished, horizontal glass. The same thing happens too with a steel ball that is touched with a weak lodestone at the point diametrically opposed to the pole of the vigorous lodestone under which it is suspended.

Similarly, if you put the point of a needle ([Physics plates], fig. 69 ) under one of the poles of the lodestone, such that it is suspended head down, and any iron bar F is presented to this head by its upper extremity, the needle will instantly leave the lodestone to attach to the bar; yet if the needle holds by its head to the pole of the lodestone, then neither the iron bar nor a weak lodestone will detach it. It would seem that the needle would attach to the one of the two which it touched at the most points, but experiments done on purpose have proved the contrary.

Another rather slight circumstance further causes a strong, armed lodestone to seem to have no remaining force: it is the excessive length of iron that one tries to lift by one of its poles. It would be easy to get certain lodestones to lift a piece of cubic iron weighing one pound; but the same lodestone would not be able to hold an iron rod of one foot in length: so that increasing the length of the suspended body is a means of diminishing the effect of the attractive property of the lodestone’s poles. That is why when you present a pole of a good lodestone on a pile of needles, small nails, or rings, the lodestone only attracts seven or eight of them end to end, and it is easy to observe that the attraction from the first nail to the second is much stronger than from the second to the third, and so forth, so that the attraction from the next-to-last to the last is extremely weak. See [Physics plates], fig. 34 .

III. On communication of the magnetic property

The lodestone can communicate directional and attractive qualities to iron, and one should consider their recipient in this manner as an authentic lodestone which can itself also transmit them to other piece of iron. A vigorous lodestone will thus give power to a weak lodestone and forever make its effects as perceptible and as powerful as those of a good lodestone.

In general, it is enough to touch or even just approach the pole of a good stone to the body to which the magnetic property is to be communicated, and at once it is magnetized. To be sure, the iron which has absorbed some of this property only for an instant of contact with the lodestone will lose it almost as soon as the two are separated; but its power can be made more lasting by leaving it longer with the lodestone, or by heating it red hot before approaching the stone, and allowing it to cool there: in that case, the part presented to the lodestone’s boreal pole will become an austral pole, and would similarly become a boreal pole if it were approached to the lodestone’s austral pole.

But since these simple means do not produce much force, more effective means are ordinarily used.

First it was discovered that iron rubbed on one of the poles of the lodestone acquires much more force than on any other part of the stone, and that the power which this pole communicates to the iron is much more considerable when it is armed than when it is in its natural state. Second, the more slowly the iron is rubbed, and the more it is pressed against the pole of the lodestone, the more of the magnetic property it absorbs. Third, it is more advantageous to magnetize the iron on a single pole of the lodestone than successively on the two poles, because the iron absorbs the magnetic property from each pole in opposite directions, with effects that cancel each other out. Fourth, a piece of iron is more easily magnetized by being rubbed uniformly and in the same direction on the lodestone’s pole along its length, than by simply rubbing it in the middle, and it is observed that the extremity that last touches the pole retains the most strength. Fifth, a piece of polished steel or else a piece of steel-clad iron absorbs more magnetic force than a piece of plain iron of the same shape; and, other things being equal, you more strongly magnetize a long, thin, and pointed piece of iron than another of completely different shape. Thus a saber, sword, or knife blade absorb much more of this property than a square of steel of the same mass which has no other points than its angles. In general, a piece of iron or steel passed over the pole of a lodestone, as we have said, never absorbs, or rather preserves, more than a determined quantity of magnetic power; and it appears that this quantity of magnetic power is determined by the length, width, and thickness of the piece of iron or steel being magnetized. Sixth, since the iron absorbs the magnetic property only along its length, it is important, in order to give it great magnetic power, that this length be fairly considerable: that is why a sword blade absorbs more magnetic property than a knife blade rubbed on the same stone. Still, there are certain proportions of thickness and length beyond which the iron absorbs less magnetic power. Here is an example: we have magnetized six iron leaves 4 inches in length and of about 1/100 inch [ sic ] thickness; their respective widths were 1, 2, 3, 4, 5, and 6 lines; each was passed three times in the same manner over the pole of an excellent lodestone, and the different weights they could lift tested. The first, which was the smallest, lifted

1 grain ¼

The 2 ^nd two lines wide, 10 ⅝

The 3 ^rd three lines wide, 7 ⅝

The 4 ^th four lines wide, 2 0

The 5 ^th five lines wide,1 ½

The 6 ^th six lines wide, 1 1/10

Here is now proof that the magnetic strength a piece of iron can absorb from a lodestone depends also on the proportion of its length. We took an iron blade 1/100 inch in thickness, 5 lines wide and 13¼ inches long; we passed it three times over the pole of a lodestone, and it could hold 25 grains; we reduced it to the length of 10 inches, and magnetized it three more times: it held 33 grains; reduced to nine inches, it held 19 grains; at 8 inches, 17 grains; at 4 inches, 1½ grains: whence it can be seen that the length must be set at 10 inches or between 10 and 13¼ so that with the given width and thickness this bar can acquire the most magnetic power.

When a blade of iron or steel of a certain width and thickness is too short to absorb much of the magnetic property by communication, it can be improved by attaching it to another, longer piece of iron of about the same width and thickness, so that the whole is about as long as is necessary so that a bar having these same dimensions could acquire the most magnetic power possible in passing it over the pole of the lodestone; then, separating the small bar from the large one, we will find its magnetic power considerably augmented. That is how we found the means of considerably augmenting the magnetic power of a piece of saber blade one foot in length, by applying it to another that was 2 feet, 7 inches, and 8 lines in length, and magnetizing them in this situation: then the small blade which could hold, when magnetized alone, only 4 ounces, 2 gros , [16] 36 grains, lifted after being separated from the large one, 7 ounces, 3 gros , 36 grains.

We must however observe that two blades thus combined do not absorb as much magnetic property as a single blade of the same length and dimensions. For we cut in two very equal parts a moderately thin iron blade and divided one of these halves into several rectangular pieces; we put together the sawn parts to each other so they could make about the same length they had had before, and fixed them in this position; we placed beside them the half of the blade that had not been cut, and magnetized them both equally. The part that had remained intact had much more magnetic force than the other, and the cut-up part absorbed less as its fragments were less contiguous to each other.

Independently of these methods for communicating the magnetic property to iron by means of the lodestone, there are others which we shall discuss later in dealing with artificial magnetism. But we cannot fail here to make it known that there are means of giving iron a very considerable magnetic power, and even increasing that of weak lodestones to the point of making them quite powerful. Mr. Knight of Magdalen College at Oxford, is the author of this discovery, which he has not yet made public. [17] Here are some examples of the great magnetic power which he communicated to steel bars which they could not obtain by magnetizing them on the best lodestones in the ordinary manner. 1. A small steel bar with eight sides, of 3 inches 7/10 in length, and weighing about a half-ounce, lifted about eleven ounces with one of its ends without being armed. 2. Another parallelepiped steel bar of 59/10 [ sic ] inch long, 4/10 inch wide, and 2/10 thick, weighing two ounces 8½ grains, lifted twenty ounces with one of its ends without being armed. 3. Another bar of the same shape and 4 inches long, armed with steel like a lodestone, the armor held in place with a silver band, the whole weighing one ounce fourteen grains, lifted four pounds with the foot of its armor. 4. A parallelepiped steel bar four inches long, 1 and 2/10 wide and 4/10 inch thick, armed at its ends with a band of copper to hold the armor, the whole weighing fourteen ounces and one scruple, [18] lifted with one of the feet of its armor fourteen pounds two and one-half ounces.

He also made an artificial lodestone with twelve bars of steel armed in the usual manner, which raised by one of the feet of its armor 23 pounds two and one-half ounces. Each of these 12 bars was a little longer than four inches, 3/10 inches wide, and 16/100 thickness; each of these blades weighed about 25 scruples, and they were placed one on top of another, so that together they formed a parallelepiped about two inches high. All these blades were tightly bound with copper bands, and carried the usual steel armor; the whole weighed 20 ounces.

The method for communicating a great magnetic power, particular to Mr. Knight, is not limited to iron and steel; he also knows how to magnetize a weak lodestone to the point of making it excellent: he presented one to the Royal Society in London which, armed, weighed seven scruples 14 grains, and which could scarcely lift two ounces. Having magnetized it at various times following his method, it lifted as much as 13 ounces. He so magnetized this weak lodestone that he made the force at its poles disappear and then substituted others more vigorous and directly opposite, such that he put the boreal pole where the austral pole naturally was, and thus with the other pole; similarly, he put the poles of a lodestone where the equator used to be, and the equator where the poles were. In a cylindrical lodestone he put a boreal pole all around the circumference of the circle that was one of its bases, and the austral pole in the center of that same circle, while the whole circumference of the other base is an austral pole, and the center a boreal pole. At will he puts the poles of a lodestone in whatever location one might wish: for example, he turns the middle of a stone into a boreal pole and the two ends into the austral pole. Finally, in a parallelepiped lodestone he puts the poles at the two ends such that the upper half of the surface is an austral pole and the lower half a boreal pole ; the upper half of the other end is a boreal pole and the lower an austral pole .

It is likely that Mr. Knight succeeds in producing all these effects by some means analogous to the one revealed to the public by Mr. Mitchell, [19] in other words with the help of artificial lodestones made with bars of tempered and polished steel, magnetized in a particular way which he calls the double contact . It is very certain that one can give to steel bars of an appropriate shape, and tempered very hard, a very considerable quantity of magnetic force. Tempered steel has the advantage over iron and soft steel that it retains much more magnetic force, although it does not so easily absorb it, and that the poles can be put at whatever distance one wishes from each other, and in the places one deems most convenient. We shall shortly explain in the section on the artificial lodestone the manner of magnetizing by means of the double contact .

Communication of magnetic virtue in no perceptible way exhausts the lodestone from which the virtue has been borrowed. Whatever the number of pieces of iron one has magnetized with a single stone, one in no way diminished its strength, even though we have seen lodestones that have given the iron more power for raising weights than they themselves had, without their strength for that appearing to be diminished.

Nor does the iron enrich itself at the expense of the lodestone, whatever force it acquires from it. For we carefully weighed a blade of polished steel and an armed lodestone, and after noting the weight of each separately, we magnetized the blade: after the operation we found the weight of these two bodies exactly the same, although we used a very precise balance.

Moreover, it is not the lodestones that lift the greatest weights that communicate the most power: experience has taught that very small lodestones, and very weak for lifting iron, can yet communicate great magnetic power. It is true that there are kinds of iron that accept none of the property from a good lodestone, whereas another kind of iron absorbs a great deal. But this truth does not appear in a more evident manner than in artificial lodestones, which for the most part communicate great power, and which nevertheless ordinarily lift little iron.

Artificial lodestone

When a piece of iron or steel is magnetized, it can communicate the magnetic property to other iron and even to lodestone itself (if it is strong enough): then it differs in no way from the lodestone as to effects; that is why it is called artificial magnet . [20] Among the methods for making artificial magnets, here is the one that was proposed as the best.

Choose several well-tempered foil blades, polished and well calibrated, so that they are equal in length, width, and thickness; they will be about six inches long, five lines wide, and one line thick, and if you want to increase their length, you will increase their other dimensions in proportion. Magnetize well each blade separately on the pole of an excellent, well-armed lodestone; prepare an armor ABCD ([Physics plates], fig. 36 ) that can contain them all pressed together, and which binds and embraces them by the knobs C and D installed about their ends. The thickness of the jamb A and B as well as those of the knobs C and D must be all the greater that there are a larger number of bars assembled. When all the bars have been placed on top of each other between the two jambs so that the poles of the same name are all on the same side, subject them in this situation by the means of the screws O , O , P , P , and the artificial magnet will be finished.

Sometimes one is content to join together several foil blades each separately magnetized, while maintaining all their length; they are kept in line by circles of copper, taking care that all their ends are on the same plane. It is on this end that you pass the blades of steel and the needles you wish to magnetize, and these sorts of artificial magnets are preferable to many natural lodestones. These artificial magnets will be all the better if they are constructed from excellent steel, well-tempered and well-polished, if they are passed over the pole of a very vigorous natural or artificial magnet, if they are of greater length, and finally if more of them are combined.

Yet we must confess that despite all these precautions, without a sufficiently strong lodestone there is no way to communicate to the steel bars that make up the artificial magnet all the magnetic power they are capable of absorbing and containing; for we must observe that a given piece of steel is capable of a determined quantity of magnetic power, beyond which it cannot acquire, or at least retain, any more. Therefore, it would be most advantageous if we could easily give steel leaves the full quantity of magnetism they can absorb: this is precisely the advantage of Mr. Mitchell’s method called the double contact , a method by which he makes artificial magnets far superior to those that can be made with the previous methods, and stronger even than the best natural lodestones. This method consists in the following.

You take twelve flat and equal steel bars, six inches long and six lines wide, and of such a thickness that they weigh only about 1¾ ounces. After filing and adjusting them, you have them heated red hot over a moderate fire (for too great or too weak a fire, would not work as well), and temper them. Near one of their ends you make a mark with a chisel or punch so as to recognize the pole that is to turn towards the north, and which is called the austral pole .

All these bars thus prepared, you place six on a table in a single straight line, in the approximate direction of the magnetic meridian, and orient them so that all the ends marked with the chisel are turned toward the north, and touch the unmarked end of the next bar; then you take a good armed lodestone and place its two poles on one of the bars such that its north pole is turned toward the marked end of the bar, which is to become the austral pole, and the austral pole of the lodestone is turned toward the end of the bar that is unmarked, and is to become a boreal pole. You slide the lodestone on both sides from one end to the other of the line formed by these six bars, and repeat the same operation three or four times, taking care to touch all of them; then, bring the lodestone back over one of these bars in the middle, you remove the two bars at the ends, and place them in the middle of the line situated as they were before, after which you again pass the stone three or four times over them, but without going this time all the way to the end of the line, because the bars which are now at the ends, and were previously in the middle, already have more power than they could absorb at the ends of the line where they now are, and they would even lose part of it if they were again passed over; and it is precisely because the bars at the ends do not absorb as much power as those in the middle that it is advised to put them back in the middle to pass them over again.

After all these operations have been carried out, it is advised to turn all the bars upside down, and touch them again on the other side, except for those at the ends which will not be touched again, for the reasons just given, but which you will bring back to the middle to touch them again after the others. Having thus communicated a little magnetism to the six steel bars, you place the six others on a table in the same manner as the preceding ones. You can see in [Physics plates] figure 72 the disposition of three of these bars A B , and the punch or chisel marks which are on their right-hand ends, where their austral pole should be. C D and E F represent the six other bars already magnetized, as we have just said, of which three are in the assemblage C D and three in E F . [21] They all touch at the top, but at the base they are separated by an inch or a little more, although at first, when they have but little power, they can be placed closer together as long as they do not touch, which they must never do.

To prevent them from touching, you can place between them a little piece of wood or any other material, as long as it is not iron.

The three magnets C D (for we can already call them that, although their power is still very weak) all have their austral pole below and on the side of the ends of the bars that are not marked, in other words those that should become the boreal pole; and the three magnets E F have their boreal pole below turned toward the ends of the marked bars. When all six are thus disposed, they will be run three or four times from one end to the other in both directions; then the end bars will be brought back to the middle to pass them again as we have said above, and they will all be turned over to do the same thing on the other surface.

If the first six bars C D , E F have been magnetized by a sufficiently vigorous lodestone, the last six will already be more strongly magnetized than the first six; that is why the first six will be placed in a straight line on a table as before, and passed again the same way with the last six until they have become even stronger; then they will be used to magnetize the second half-dozen in the same manner, and this operation will be repeated until these bars no longer appear to acquire any [magnetic] property by repeated contacts.

Each of these six bars, when it has been well tempered and magnetized in the manner we have just explained, will be able to lift a piece of iron of one pound or more with one of its poles (provided it be of a suitable shape), and six of these bars once well magnetized and used in the manner we have just taught, will completely magnetize six new bars by passing them only three or four times from one end to the other, except for those on the ends, which must always be passed again after being brought to the middle.

In all these operations one is often obliged to separate or reconnect the iron bars that make up the two packs C D , E F , as well as the six that make up the line A B . Now, as two magnets that have like-named poles on the same side always weaken each other when they come into contact, it is absolutely necessary (and one must take very great care in every instance) never to place two at once on the same side C D or E F ; but they should be placed one by one on each side, making them touch their entire length, or else by putting their lower ends on the line of the bars to be magnetized while they are touching at the upper ends; and the same thing shall be observed when withdrawing them, that is, one by one on each side. It will be quicker to assemble all six in a bundle by taking them one by one at the same time on each side; and moving them onto the line of bars, they will be divided into two bundles, as we have shown; but one will take care to separate them at the bottom before they are on the bar, for at that moment they would weaken. Besides, if they should happen to be weakened by this accident, they could be magnetized by passing them again with the six others in the manner which we have shown.

The same precautions must be used to preserve these magnetized bars. That is why you will have a suitable box in which you will place just so two pieces of iron about one inch long (which is about the thickness of the six steel bars) perpendicularly one to the other, and at a distance of six inches from the outside of each; these iron pieces will be about a quarter-inch square and well-polished on the sides: you will place beside them, and in contact, the twelve steel bars, six on one side and six on the other, the six on one side with their north pole toward one end of the box, and the six on the other with their south pole toward that end. You must take care never to put them in nor take them out all at once on either side, for you would demagnetize them; but you will place them one at a time on each side, such that their efforts are constantly in counter-balance. It is an observation one must always make, never to leave two or more together with their same-named pole on the same side, or else they would not fail to lose their [magnetic] property.

The magnetic property which you communicate to a piece of iron or steel remains there as long as those bodies are not exposed to any violent action that can dissipate it. There are nonetheless rather small circumstances that can destroy in very little time the magnetism of the best-magnetized steel. We shall list the principal ones below.

First, when you have magnetized a piece of iron on a vigorous lodestone, if you were to pass it on the same pole of a weaker lodestone, it would lose much of its virtue, and preserve only enough as a weak lodestone over which it had most recently passed could have given it. Second, when you pass an iron or steel blade over the same pole of the lodestone on which it has already been magnetized, but in the opposite direction to the first, the magnetic property of the blade is at once dissipated, and can be re-established only by continuing to pass the blade over the same pole in the last direction; but the poles will be changed at each end, and it will be very difficult to communicate to it as much magnetic power as it had at first.

3. It is essential to be sure to touch the poles of the lodestone with the piece of iron you wish to magnetize, and not be content to approach it within a small distance, not only because it is the best way to communicate much magnetic virtue to it, but because the magnetic material distributes itself in the iron in a single and same direction. Here is an experiment that proves the necessity of contact between the iron and the armor of the lodestone for the communication to be perfect. If you pass a compass needle from one pole to the other of a lodestone, having it touch in succession the two sheaths of the armor, it will acquire magnetic power, and orient itself north and south, as we know. But if, after examining its orientation, you pass it a second time over the lodestone in the same direction you did at first, only with the difference that instead of touching the sheaths of the armature, you only bring it close, even the closest possible, its magnetic power will at once diminish, and it will acquire a different one, but with an orienting power exactly opposite the first. And if you continue to magnetize it in the same direction, beginning again to touch the sheaths of armor, this second magnetic property will disappear, and it will assume another one with its initial orientation, and you will destroy its magnetism and orientation in this manner as many times as you might wish.

4. To preserve well the magnetic property you have communicated to a piece of iron, it must be protected from any violent shock; for any sudden and irregular shock destroys the magnetism. We magnetized a steel blade on an excellent lodestone, and after recognizing its attractive property, which was very strong, we pounded it for some time on an anvil: it had soon lost all its power, except that it could indeed lift a few iron filings, as all forged iron can do; but it was never able to life the smallest needle. The same thing would have occurred by throwing it several times onto a square of marble.

5. The action of fire also destroys a large portion of the magnetic property which has been communicated. After well magnetizing an iron blade, it is heated red-hot in the fire of a forge until it turns white; when it was presented at this temperature to some iron filings, it attracted none of them; but it recovered the magnetism as it cooled. Yet when an iron blade that was already red was magnetized, it attracted iron filings, and this attraction was more intense after the blade had been allowed to cool.

6. The action of folding or twisting a piece of magnetized iron also causes it to lose its magnetic property. We magnetized a piece of iron rod so that it oriented itself sharply, following the magnetic meridian; then we bent it into a ring, and we found that it no longer had any orientation in this shape. We brought it back to its first state, but all this violence had taken away its magnetic property, such that it longer oriented itself. We conjectured that the two poles had acted on each other at the point of contact, and had cancelled each other out; we therefore again magnetized the same iron rod and several similar ones, and made them into imperfect rings. We observed that they too had lost their magnetic property in this new shape, and that they recovered it only when they were made straight. This experiment always succeeds when the iron rod is well and duly curved, and especially if you make it turn several times into a spiral around a cylinder: for if the least of its parts is not curved with violence, it will preserve its magnetism. The same will occur with a magnetized iron rod which you bend first in two, and of which you twist the two halves about each other: so it appears that the magnetism is destroyed by the violence imparted to the iron in all these cases, and from the derangement caused in its parts, as one can easily see by means of the microscope.

Here is an experiment that confirms this truth, and makes it apparent that the disturbance caused to the parts by the iron destroys the magnetism. We placed some iron filings in a well-dried glass tube, and pressed it down carefully; we gently magnetized it with a good armed lodestone, and the tube attracted the filings spread on a table; but as soon as we had shaken the tube, and changed the respective positions of the filings, the magnetic property disappeared.

On iron magnetized without ever touching a magnet

A lodestone, or an artificial magnet, is not always needed to communicate magnetic property to iron and steel: these bodies sometimes take on magnetism naturally; we magnetize them sometimes by different means without it being necessary to borrow the aid of any magnet.

First, any piece of iron oblong in shape, which remains for a long time in a vertical position, becomes a magnet that much more perfect as it has remained longer in this position. That is how the crosses of the bell towers of Chartres , Delft , Marseille , etc., have become such perfect magnets that they have almost lost their metallic quality, and attract and exert all the effects of the best lodestones. Moreover, the magnetic property they have thus contracted over time has remained fixed and constant, and manifests itself in all sorts of situations. To see this, you need only set vertically on a cork C a piece of iron a b ([Physics plates,] figure 54 ) which has long remained in a vertical position, and place the whole in water. If you approach the boreal pole B of a magnet to the upper end a of this piece of iron, the iron will be attracted, but it will be repelled if you present the other pole A of the magnet; likewise if you approach the pole A of the lower end b of the iron, it will be attracted, and repelled if you approach the pole B of the magnet.

In the second place, shovels and tongs, the iron bars of windows, and in general all pieces of iron that remain for a long time in a position perpendicular to the horizon, acquire a more or less permanent magnetic property, depending on the time they have remained in that position; and the upper part of these bars always becomes an austral pole, whereas the bottom is a boreal pole.

3. There are certain circumstances in which lightning communicates great magnetic power to iron. It once struck a room in which there was a case filled with steel knives and forks destined to go to sea; the lightning entered through the southern corner of the room right where this case was; several knives and forks were melted and broken; others that remained intact were very vigorously magnetized, and became capable of lifting large nails and iron rings, and this magnetic property was so strongly imprinted on them that it was not dissipated by heating them to red hot.

4. The same iron bar can acquire fixed or variable magnetic poles without touching the magnet, which one will discover easily by means of a needle magnetized in this manner. A magnetized needle, freely mobile on its pivot, is approached by an iron bar that has never touched the magnet nor remained long in a vertical position; this iron bar is held horizontally, and the needle remains immobile whichever pole is presented to it; as soon as the bar is presented in a vertical position, at once its upper end strongly attracts (in this northern hemisphere of the earth) the boreal end of the needle, and the lower part of the bar attracts the south of the needle ([Physics plates,] figure 55 ). [22] But if the bar is reversed, such that its upper part is the same that was at the bottom in the previous case, the north of the needle will still be constantly attracted by the upper end of the bar, and the south by the lower end; whence it is evident that the vertical position determines the poles of an iron bar: to wit, the upper edge is always (in our hemisphere) an austral pole, and the lower one a boreal pole; and as each end of the bar can be placed up or down, it is clear that the poles it acquires by this method are variable. To give an iron bar fixed poles, do this: heat it to red hot, and let it cool while holding it in the meridian plane: then the end facing north becomes a constant boreal pole, and the one which cools to the south becomes a constant austral pole. But for this experiment to succeed, there must be a certain proportion between the thickness of the bar and its length; for example, a bar of 1/5 inch diameter must be at least 30 inches long to acquire fixed poles by this method, and a bar 30 inches long must have only 1/5 inch diameter; for if it were thicker, it would have only variable poles.

5. We have previously seen that a strong and prompt percussion on a piece of magnetized iron is capable of destroying its magnetic property; a similar percussion on a piece of iron which has never touched the magnet is capable of giving it poles. We placed a bar of soft iron, long and thin, on a large anvil, and in the meridian plane, and struck with a hammer the end that was turned toward the north: at once it became a boreal pole. We similarly struck the other end, which became an austral pole. One must always observe in these sorts of experiments that the length of the bar be proportionate to its thickness, without which they do not succeed. This effect, moreover, which one produces with a hammer, also occurs by filing or sawing the bar at one of its ends.

6. The steel tools that serve to cut or pierce iron become magnetized by the work, especially as they heat up, so there are some that can lift small iron nails. These tools have almost no strength after being tempered, but when after being heated again they are filed and worn down, they then acquire much power which nevertheless diminishes when they cool down. Pieces of steel that end in a point become much more magnetized than those that end in a wide, flat tongue: thus a steel punch attracts more at its point than a chisel or an ordinary knife. The longer the punches are, the more [magnetic] property they acquire, so that a punch one inch long and 9 lines in diameter attracts much less than a drill of 3 to 4 inches and 1½ lines in diameter.

We have observed that the attractive property of all bodies magnetized in this manner was much stronger when its effect was tried on an anvil or some other large piece of iron; so that by all appearances, small nails that have become artificial magnets by contact with the anvil presented to punches their differently-named poles, which made the attraction stronger than when they were on any other body, or no longer had any polar property.

7. One can also magnetize very well a soft and flexible piece of iron, again of a length proportionate to its thickness, by breaking it on one or the other of its ends by bending it back and forth. This is how we magnetized a piece of very flexible iron rod, two and a half feet long, and thick as a little finger; we tightened it in a vice five inches from its end, and after bending it back and forth we broke it: each of its ends attracted a small tack at the break point. We put the longest part back in the vice and held it at a half inch from the break, and bent it back and forth without breaking it, and we found its attractive power considerably increased at the point of the break; we bent it in this way eight different times up to the middle, and it could lift four tacks; but when we continued bending it beyond the middle toward the other end, its power diminished at the point of the break, and it attracted, to the contrary, by the opposite end, until, after it had been bent several times up to this latter end, it lifted four tacks with it, while it could barely lift a few iron filings by the end where it had been broken.

If you bend a piece of iron in the middle, it will acquire almost no magnetic property; it you bend it at equal distances from the middle, each of its extremities will be magnetized, but more weakly than if it had been bent only on one side.

8. Finally, Mr. Marcel, of the Royal Society of London, has found a means of communicating the magnetic property to pieces of steel which is again independent of the magnet. [23]

This means consists of putting pieces of steel on a well-polished anvil, and rubbing them lengthwise, and always in the same direction, with a fat vertical iron bar the lower end of which is rounded and well-polished; by repeating this rubbing a great many times on all the faces of the piece of steel to be magnetized, it acquires as much magnetic power as if it had been contacted by the best magnet. In this way he magnetized compass needles, steel blades for making artificial magnets, and knives that could hold one and three-fourths ounces.

In the pieces of steel which one magnetizes in this way, the end by which the rubbing begins is always facing north; and the end by which the rubbing ends faces south, whatever be the position of the steel on the anvil.

This experiment succeeds, moreover, much better when the piece of iron or steel to be magnetized by this method is in the direction of the magnetic meridian, a little angled toward the north, and especially between two fat iron bars long enough to contain and counterbalance the effort of the magnetic effluvia being impressed upon the piece of steel.

This entire article was given to us by M. Lemonnier, physician, of the Royal Academies of Science of Paris and Berlin, who has made a particular study of the lodestone with great success. On the cause and properties of the lodestone, see Magnetism.

1. En roche . “Iron consists of an earth, salt, and sulphur, but all impure, ill mixed, and digested” (Chambers, Cyclopaedia, 1741).

2. Vertu . “Efficacity, force, vigor, property.” Dictionnaire de l’Académie française (1694). In this article, vertu will sometimes be translated as property , as in this instance, sometimes as power or force .

3. 540–537 BCE.

4. “What is more impassive than the stiffness of stone? And yet we see that she has endowed the magnet with senses and hands. What is more recalcitrant than the hardness of iron? We see that she has bestowed on it feet and instincts. For iron is attracted by the magnet, and the substance that vanquishes all other things rushes into a kind of vacuum and, as it approaches the magnet, it leaps towards it and is held by it and clasped in its embrace.” Pliny the Elder, Natural History , trans. D. E. Eichholz, Loeb Classics, 1962, Vol. X, p. 101.

5. “I have seen iron rings / from Samothrace even leaping around / and, at the same time, iron filings moving / frantically inside bronze bowls [...], ” De rerum naturae , Book VI, 1044–46.

6. Vertu directive , by which he means the lodestone’s response to the earth’s magnetic field.

7. The references here are to the natural philosopher Robert Boyle (1627-1691) and the physicist Gowin Knight (1713-1772).

8. Sphère d’activité , what would now be called an electromagnetic field.

9. Étienne François Dutour de Salvert (1711–1789), won a prize from the Académie des Sciences in 1746 for his Mémoire sur l'attraction de l'aimant [Paper on the attraction of the lodestone] and would in 1768 publish Recherches sur les différents mouvements de la matière électrique [Research on the different motions of electric matter].

10. 120 seconds.

11. Pieter van Musschenbroek (1692-1761), Elementa Physicæ [Elements of physics] (1726; link is to 1745 edition published in Naples); translated into French as Essai de physique (1739; link is to an edition published in Leiden in 1751).

12. In Paris weights, the grain was equal to .053 grams.

13. John Michell (1724–1793), A Treatise of Artificial Magnets , 1750.

14. Armure “is said [...] in speaking of the lodestone, of two pieces of iron that are placed on the poles of that stone, and bound tightly with a small belt of metal. This armor very considerably increased the power of the lodestone.” ( Dictionnaire de Trévoux .)

15. A line is one-twelfth of an inch.

16. A gros is about ⅛ ounce.

17. “An account of some magnetical experiments, shewed before the Royal Society, by Mr. Gowan Knight, on Thursday the 15 ^th of November, 1744,” Philosophical Transactions , vol. 43, no. 474, article VIII.

18. Twenty-four grains.

19. See note 13.

20. The term aimant in the original, which can refer to either lodestone or magnet, does not change; but as the article now begins to speak less about stones than about magnetized steel, the term “magnet” is being substituted where appropriate to recognize that difference in English.

21. In some places, as here, the plates have not been completed in a way to correspond accurately with this description.

22. Figure 55 seems to have been overlooked when the various images were fitted into plate IV of the Physics series.

23. Arnold Marcel (1672–1748) was not a fellow of the Royal Society, but knew someone (John Thomas Desaguliers) who was, and who had published for him An abstract of a letter, written in Dutch, to the illustrious Royal Society of London [...] communicated by the Revd. Dr. J. T. Desaguliers, F. R. S. ( Philosophical Transactions , vol. 37, no. 423 (1731), pp. 294–298.