|
The Dibner Library of the History of Science and Technology SMITHSONIAN INSTITUTION LIBRARIES |
||||
[ 21 ]
II. Experimental Researches in Electricity. — Twentieth Series. By Michael Faraday, Esq., D.C.L. F.R.S., Fullerian Prof. Chem. Royal Institu- tion, Foreign Associate of the Acad. Sciences, Paris, Cor. Memb. Royal and Imp. Acadd. of Sciences, Petersburgh, Florence, Copenhagen, Berlin, Gbttingen, Modena, Stockholm, 8$c. fyc.
Received December 6, — Read December 18, 1845.
§ 27- On new magnetic actions, and on the magnetic condition of all matter*. ^[ i, Apparatus required. *f[ ii. Action of magnets on heavy glass. ^[ iii. Action of magnets on other substances acting magnetically on light. ^[ iv. Action of magnets on the metals generally.
2243. 1 HE contents of the last series of these researches were, I think, sufficient to justify the statement, that a new magnetic condition (i. e. one new to our know- ledge) had been impressed on matter by subjecting it to the action of magnetic and electric forces (2227.) ; which new condition was made manifest by the powers of action which the matter had acquired over light. The phenomena now to be described are altogether different in their nature ; and they prove, not only a mag- netic condition of the substances referred to unknown to us before, but also of many
* My friend Mr. Wheatstone has this day called my attention to a paper by M. Becquerel, " On the magnetic actions excited in all bodies by the influence of very energetic magnets," read to the Academy of Sciences on the 27th of September 1827, and published in the Annales de Chimie, xxxvi. p. 337. It relates to the action of the magnet on a magnetic needle, on soft iron, on the deutoxide and tritoxide of iron, on the tritoxide alone, and on a needle of wood. The author observed, and quotes Coulomb as having also observed, that a needle of wood, under certain conditions, pointed across the magnetic curves ; and he also states the striking fact that he had found a needle of wood place itself parallel to the wires of a galvanometer. These effects, however, he refers to a degree of magnetism less than that of the tritoxide of iron, but the same in character, for the bodies take the same position. The polarity of steel and iron is stated to be in the direction of the length of the substance, but that of tritoxide of iron, wood and gum-lac, most frequently in the direction of the width, and always when one magnetic pole is employed. " This difference of effect, which establishes a line of demarcation between these two species of phenomena, is due to this, that the magnetism being very feeble in the tritoxide of iron, wood, &c, we may neglect the reaction of the body on itself, and therefore the direct action of the bar ought to overrule it."
As the paper does not refer the phenomena of wood and gum-lac to an elementary repulsive action, nor show that they are common to an immense class of bodies, nor distinguish this class, which I have called diamagnetic, from the magnetic class ; and, as it makes all magnetic action of one kind, whereas I show that there are two kinds of such action, as distinct from each other as positive and negative electric action are in their way, so I do not think I need alter a word or the date of that which I have written ; but am most glad here to acknow- ledge M. Becquerel' s important facts and labours in reference to this subject.— M. F. Dec. 5, 1845.
22 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
others, including a vast number of opake and metallic bodies, and perhaps all except the magnetic metals and their compounds : and they also, through that con- dition, present us with the means of undertaking the correlation of magnetic pheno- mena, and perhaps the construction of a theory of general magnetic action founded on simple fundamental principles.
2244. The whole matter is so new, and the phenomena so varied and general, that I must, with every desire to be brief, describe much which at last will be found to concentrate under simple principles of action. Still, in the present state of our knowledge, such is the only method by which I can make these principles and their results sufficiently manifest.
^[ i. Apparatus required.
2245. The effects to be described require magnetic apparatus of great power, and under perfect command. Both these points are obtained by the use of electro- magnets, which can be raised to a degree of force far beyond that of natural or steel magnets ; and further, can be suddenly altogether deprived of power, or made ener- getic to the highest degree, without the slightest alteration of the arrangement, or of any other circumstance belonging to an experiment.
2246. One of the electro-magnets which I use is that already described under the term Woolwich helix (2192.). The soft iron core belonging to it is twenty-eight inches in length and 2*5 inches in diameter. When thrown into action by ten pair of Grove's plates, either end will sustain one or two half-hundred weights hanging to it. The magnet can be placed either in the vertical or the horizontal position. The iron core is a cylinder with flat ends, but I have had a cone of iron made, two inches in diameter at the base and one inch in height, and this placed at the end of the core, forms a conical termination to it, when required.
2247- Another magnet which I have had made has the horse-shoe form. The bar of iron is forty-six inches in length and 3*75 inches in diameter, and is so bent that the extremities forming the poles are six inches from each other; 522 feet of copper wire 0"17 of an inch in diameter and covered with tape, are wound round the two straight parts of the bar, forming two coils on these parts, each sixteen inches in length, and composed of three layers of wire : the poles are, of course, six inches apart, the ends are planed true, and against these move two short bars of soft iron, 7 inches long and 2^ by 1 inch thick, which can be adjusted by screws, and held at any distance less than six inches from each other. The ends of these bars form the opposite poles of contrary name ; the magnetic field between them can be made of greater or smaller extent, and the intensity of the lines of magnetic force be pro- portionately varied.
2248. For the suspension of substances between and near the poles of these magnets, I occasionally used a glass jar, with a plate and sliding wire at the top. Six or eight lengths of cocoon silk being equally stretched, were made into one thread
ON NEW MAGNETIC ACTIONS. 23
and attached, at the upper end, to the sliding rod, and at the lower end to a stirrup of paper, in which anything to be experimented on could be sustained.
2249. Another very useful mode of suspension was to attach one end of a fine thread, six feet long, to an adjustible arm near the ceiling of the room, and termina- ting at the lower end by a little ring of copper wire ; any substance to be suspended could be held in a simple cradle of fine copper wire having eight or ten inches of the wire prolonged upward ; this, being bent into a hook at the superior extremity, gave the means of attachment to the ring. The height of the suspended substance could be varied at pleasure, by bending any part of the wire at the instant into the hook form. A glass cylinder placed between the magnetic poles was quite sufficient to keep the suspended substance free from any motion, due to the agitation of the air.
2250. It is necessary, before entering upon an experimental investigation with such an apparatus, to be aware of the effect of any magnetism which the bodies used may possess; the power of the apparatus to make manifest such magnetism is so great, that it is difficult on that account to find writing-paper fit for the stirrup above men- tioned. Before therefore any experiments are instituted, it must be ascertained that the suspending apparatus employed does not point, i. e. does not take up a position parallel to the line joining the magnetic poles, by virtue of the magnetic force. When copper suspensions are employed, a peculiar effect is produced (2309.), but when understood, as it will be hereafter, it does not interfere with the results of ex- periment. The wire should be fine, not magnetic as iron, and the form of the sus- pending cradle should not be elongated horizontally, but be round or square as to its general dimensions, in that direction.
2251. The substances to be experimented with should be carefully examined, and rejected if not found free from magnetism. Their state is easily ascertained ; for, if magnetic, they will either be attracted to the one or the other pole of the great magnet, or else point between them. No examination by smaller magnets, or by a magnetic needle, is sufficient for this purpose.
2252. I shall have such frequent occasion to refer to two chief directions of position across the magnetic field, that to avoid periphrasis, I will here ask leave to use a _, term or two, conditionally. One of these directions is _^
that from pole to pole, or along the line of magnetic force ; JL
I will call it the axial direction : the other is the direction
perpendicular to this, and across the line of magnetic force ; and for the time, and as
respects the space between the poles, I will call it the ^equatorial direction. Other
terms that I may use, I hope will explain themselves.
24 Dll. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
^[ ii. Action of magnets on heavy glass.
2253. The bar of silicated borate of lead, or heavy glass already described as the substance in which magnetic forces were first made effectually to bear on a ray of light (2152.), and which is two inches long, and about 0*5 of an inch wide and thick, was suspended centrally between the magnetic poles (2247-), and left until the effect of torsion was over. The magnet was then thrown into action by making contact at the voltaic battery: immediately the bar moved, turning round its point of sus- pension, into a position across the magnetic curve or line of force, and after a few vibrations took up its place of rest there. On being displaced by hand from this position, it returned to it, and this occurred many times in succession.
2254. Either end of the bar indifferently went to either side of the axial line. The determining circumstance was simply inclination of the bar one way or the other to the axial line, at the beginning of the experiment. If a particular or marked end of the bar were on one side of the magnetic, or axial line, when the magnet was rendered active that end went further outwards, until the bar had taken up the equa- torial position.
2255. Neither did any change in the magnetism of the poles, by change in the di- rection of the electric current, cause any difference in this respect. The bar went by the shortest course to the equatorial position.
2256. The power which urged the bar into this position was so thoroughly under command, that if the bar were swinging it could easily be hastened in its course into this position, or arrested as it was passing from it, by seasonable contacts at the voltaic battery.
2257- There are two positions of equilibrium for the bar ; one stable, the other unstable. When in the direction of the axis or magnetic line of force, the comple- tion of the electric communication causes no change of place ; but if it be the least oblique to this position, then the obliquity increases until the bar arrives at the equa- torial position ; or if the bar be originally in the equatorial position, then the mag- netism causes no further changes, but retains it there (2298. 2299. 2384.).
2258. Here then we have a magnetic bar which points east and west, in relation to north and south poles, i. e. points perpendicularly to the lines of magnetic force.
2259. If the bar be adjusted so that its point of suspension, being in the axial line, is not equidistant from the poles, but near to one of them, then the magnetism again makes the bar take up a position perpendicular to the magnetic lines of force ; either end of the bar being on the one side of the axial line, or the other, at pleasure. But at the same time there is another effect, for at the moment of completing the electric contact, the centre of gravity of the bar recedes from the pole and remains repelled from it as long as the magnet is retained excited. On allowing the mag- netism to pass away, the bar returns to the place due to it by its gravity.
2260. Precisely the same effect takes place at the other pole of the magnet. Either of them is able to repel the bar, whatever its position may be, and at the same time the bar is made to assume a position, at right angles, to the line of magnetic force
ACTION OF MAGNETS ON HEAVY GLASS. 25
2261. If the bar be equidistant from the two poles, and in the axial line, then no repulsive effect is or can be observed.
2262. But preserving the point of suspension in the equatorial line, i.e. equidistant from the two poles, and removing it a little on one side or the other of the axial line (2252.), then another effect is brought forth. The bar points as before across the magnetic line of force, but at the same time it recedes from the axial line, increasing its distance from it, and this new position is retained as long as the magnetism con- tinues, and is quitted with its cessation.
2263. Instead of two magnetic poles, a single pole may be used, and that either in a vertical or a horizontal position. The effects are in perfect accordance with those described above ; for the bar, when near the pole, is repelled from it in the direction of the line of magnetic force, and at the same time it moves into a position perpen- dicular to the direction of the magnetic lines passing through it. When the magnet is vertical (2246.) and the bar by its side, this action makes the bar a tangent to the curve of its surface.
2264. To produce these effects, of pointing across the magnetic curves, the form of the heavy glass must be long ; a cube, or a fragment approaching roundness in form, will not point, but a long piece will. Two or three rounded pieces or cubes, placed side by side in a paper tray, so as to form an oblong accumulation, will also point.
2265. Portions, however, of any form, are repelled: so if two pieces be hung up at once in the axial line, one near each pole, they are repelled by their respective poles, and approach, seeming to attract each other. Or if two pieces be hung up in the equatorial line, one on each side of the axis, then they both recede from the axis, seeming to repel each other.
2266. From the little that has been said, it is evident that the bar presents in its motion a complicated result of the force exerted by the magnetic power over the heavy glass, and that, when cubes or spheres are employed, a much simpler indica- tion of the effect may be obtained. Accordingly, when a cube was thus used with the two poles, the effect was repulsion or recession from either pole, and also reces- sion from the magnetic axis on either side.
2267- So, the indicating particle would move, either along the magnetic curves, or across them ; and it would do this either in one direction or the other ; the only con- stant point being, that its tendency was to move from stronger to weaker places of magnetic force.
2268. This appeared much more simply in the case of a single magnetic pole, for then the tendency of the indicating cube or sphere was to move outwards, in the di- rection of the magnetic lines of force. The appearance was remarkably like a case of weak electric repulsion.
2269. The cause of the pointing of the bar, or any oblong arrangement of the heavy glass, is now evident. It is merely a result of the tendency of the particles to move outwards, or into the positions of weakest magnetic action. The joint ex-
MDCCCXLVI. E
2G DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
ertion of the action of all the particles brings the mass into the position, which, by experiment, is found to belong to it.
22/0. When one or two magnetic poles are active at once, the courses described by particles of heavy glass free to move, form a set of lines or curves, which I may have occasion hereafter to refer to ; and as I have called air, glass, water, &c. diamag- netics (2149.), so I will distinguish these lines by the term diamagnetic curves, both in relation to, and contradistinction from, the lines called magnetic curves.
2271. When the bar of heavy glass is immersed in water, alcohol, or ether, con- tained in a vessel between the poles, all the preceding effects occur ; the bar points and the cube recedes exactly in the same manner as in air.
2272. The effects equally occur in vessels of wood, stone, earth, copper, lead, silver, or any of those substances which belong to the diamagnetic class (2149.).
22/3. I have obtained the same equatorial direction and motions of the heavy glass bar as those just described, but in a very feeble degree, by the use of a good common steel horse-shoe magnet (2 157.). I have not obtained them by the use of the helices (2191. 2192.) without the iron cores.
2274. Here therefore we have magnetic repulsion without polarity, i. e. without reference to a particular pole of the magnet, for either pole will repel the substance, and both poles will repel it at once (2262.). The heavy glass, though subject to magnetic action, cannot be considered as magnetic, in the usual acceptation of that term, or as iron, nickel, cobalt, and their compounds. It presents to us, under these circumstances, a magnetic property new to our knowledge ; and though the pheno- mena are very different in their nature and character to those presented by the action of the heavy glass on light (2152.), still they appear to be dependent on, or connected with, the same condition of the glass as made it then effective, and therefore, with those phenomena, prove the reality of this new condition.
^[ iii. Action of magnets on other substances acting magnetically un light.
2275. We may now pass from heavy glass to the examination of the other sub- stances, which, when under the power of magnetic or electric forces, are able to affect and rotate a polarized ray (2 173.), and may also easily extend the investigation to bodies which, from their irregularity of form, imperfect transparency, or actual opacity, could not be examined by a polarized ray, for here we have no difficulty in the application of the test to all such substances.
2276. The property of being thus repelled and affected by magnetic poles, was soon found not to be peculiar to heavy glass. Borate of lead, flint-glass, and crown-glass set in the same manner equatorially, and were repelled when near to the poles, though not to the same degree as the heavy glass.
2277- Amongst substances which could not be subjected to the examination by light, phosphorus in the form of a cylinder presented the phenomena very well ; I think as powerfully as heavy glass, if not more so. A cylinder of sulphur, and a long
ACTION OF MAGNETS ON DIAMAGNETIC BODIES GENERALLY.
27
piece of thick India rubber, neither being- magnetic after the ordinary fashion, were well directed and repelled.
2278. Crystalline bodies were equally obedient, whether taken from the single or double refracting class (2237.)- Prisms of quartz, calcareous spar, nitre and sulphate of soda, all pointed well, and were repelled.
2279. I then proceeded to subject a great number of bodies, taken from every class, to the magnetic forces, and will, to illustrate the variety in the nature of the sub- stances, give a comparatively short list of crystalline, amorphous, liquid and organic bodies below. When the bodies were fluids, I inclosed them in thin glass tubes. Flint-glass points equatorially, but if the tube be of very thin glass, this effect is found to be small when the tube is experimented with alone; afterwards, when it is filled with liquid and examined, the effect is such that there is no fear of mistaking that due to the glass for that of the fluid. The
tubes must not be closed with cork, sealing-wax, or
any ordinary substance taken at random, for these f
are generally magnetic (2285.). I have usually so V^
shaped them in the making, and drawn them off at
the neck, as to leave the aperture on one side, so that when filled with liquid they re- quired no closing.
2280. Rock crystal.
Sulphate of lime.
Sulphate of baryta.
Sulphate of soda.
Sulphate of potassa.
Sulphate of magnesia.
Alum.
Muriate of ammonia.
Chloride of lead.
Chloride of sodium.
Nitrate of potassa.
Nitrate of lead.
Carbonate of soda.
Iceland spar.
Acetate of lead.
Tartrate of potash and antimony.
Tartrate of potash and soda.
Tartaric acid.
Citric acid.
Water.
Alcohol.
Ether.
Nitric acid.
Sulphuric acid.
Muriatic acid.
Solutions of various alkaline and
earthy salts. Glass. Litharge. White arsenic. Iodine. Phosphorus. Sulphur. Resin. Spermaceti. Caffeine. Cinchonia. Margaric acid. Wax from shell-lac. Sealing-wax. Olive-oil. Oil of turpentine. Jet. Caoutchouc.
E 2
28 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
Sugar. Beef, dried.
Starch. Blood, fresh.
Gum-arabic. Blood, dried.
Wood. Leather.
Ivory. Apple.
Mutton, dried. Bread.
Beef, fresh.
2281. It is curious to see such a list as this of bodies presenting on a sudden this remarkable property, and it is strange to find a piece of wood, or beef, or apple, obe- dient to or repelled by a magnet. If a man could be suspended, with sufficient deli- cacy, after the manner of Dufay, and placed in the magnetic field, he would point equatorially ; for all the substances of which he is formed, including the blood, pos- sess this property.
2282. The setting equatorially depends upon the form of the body, and the diver- sity of form presented by the different substances in the list was very great ; still the general result, that elongation in one direction was sufficient to make them take up an equatorial position, was established. It was not difficult to perceive that compara- tively large masses would point as readily as small ones, because in larger masses more lines of magnetic force would bear in their action on the body, and this was proved to be the case. Neither was it long before it evidently appeared that the form of a plate or a ring was quite as good as that of a cylinder or a prism ; and in prac- tice it was found that plates and flat rings of wood, spermaceti, sulphur, &c, if suspended in the right direction, took up the equatorial position very well. If a plate or ring of heavy glass could be floated in water, so as to be free to move in every direction, and were in that condition subject to magnetic forces diminishing in inten- sity, it would immediately set itself equatorially, and if its centre coincided with the axis of magnetic power, would remain there ; but if its centre were out of this line, it would then, perhaps, gradually pass off" from this axis in the plane of the equator, and go out from between the poles.
2283. I do not find that division of the substance has any distinct influence on the effects. A piece of Iceland spar was observed, as to the degree of force with which it set equatorially ; it was then broken into six or eight fragments, put into a glass tube and tried again; as well as I could ascertain, the effect was the same. By a second operation, the calcareous spar was reduced into coarse particles ; afterwards to a coarse powder, and ultimately to a fine powder : being examined as to the equa- torial set each time, I could perceive no difference in the effect, until the very last, when I thought there might be a slight diminution of the tendency, but if so, it was almost insensible. I made the same experiment on silica with the same result, of no diminution of power. In reference to this point I may observe, that starch and other bodies in fine powder exhibited the effect very well.
2284. It would require very nice experiments and great care to ascertain the
ACTION OF MAGNETS ON DIAMAGNETIC METALS. 29
specific degree of this power of magnetic action possessed by different bodies, and I have made very little progress in that part of the subject. Heavy glass stands above flint-glass, and the latter above plate-glass. Water is beneath all these, and I think alcohol is below water, and ether below alcohol. The borate of lead is I think as high as heavy glass, if not above it, and phosphorus is probably at the head of all the substances just named. I verified the equatorial set of phosphorus between the poles of a common magnet (2273.).
2285. I was much impressed by the fact that blood was not magnetic (2280.), nor any of the specimens tried of red muscular fibre of beef or mutton. This was the more striking, because, as will be seen hereafter, iron is always and in almost all states magnetic. But in respect to this point it may be observed, that the ordinary magnetic property of matter and this new property are in their effects opposed to each other ; and that when this property is strong it may overcome a very slight degree of ordinary magnetic force, just as also a certain amount of the magnetic property may oppose and effectually hide the presence of this force (2422.). It is this circumstance which makes it so necessary to be careful in examining the magnetic condition of the bodies in the first instance (2250.). The following list of a few substances, which were found slightly magnetic, will illustrate this point : — Paper, sealing-wax, china ink, Berlin porcelain, silkworm-gut, asbestos, fluor-spar, red lead, vermilion, peroxide of lead, sulphate of zinc, tourmaline, plumbago, shell-lac, charcoal. In some of these cases the magnetism was generally diffused through the body, in other cases it was limited to a particular part.
2286. Having arrived at this point, I may observe, that we can now have no diffi- culty in admitting that the phenomena abundantly establish the existence of a magnetic property in matter, new to our knowledge. Not the least interesting of the consequences that flow from it, is the manner in which it disposes of the assertion which has sometimes been made, that all bodies are magnetic. . Those who hold this view, mean that all bodies are magnetic as iron is, and say that they point be- tween the poles. The new facts give not a mere negative to this statement, but something beyond, namely, an affirmative as to the existence of forces in all ordinarv bodies, directly the opposite of those existing in magnetic bodies, for whereas those practically produce attraction, these produce repulsion ; those set a body in the axial direction, but these make it take up an equatorial position : and the facts with regard to bodies generally, are exactly the reverse of those which the view quoted indicates.
^[ iv. Action of magnets on metals generally.
2287- The metals, as a class, stand amongst bodies having a high and distinct interest in relation both to magnetic and electric forces, and might at first well be expected to present some peculiar phenomena, in relation to the striking property found to be possessed in common by so large a number of substances, so varied in their general characters. As yet no distinction associated with conduction or non-
30 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
conduction, transparent or opake, solid or liquid, crystalline or amorphous, whole or broken, has presented itself; whether the metals, distinct as they are as a class, Mould fall into the great generalization, or whether at last a separation would occur, was to me a point of the highest interest.
2288. That the metals, iron, nickel and cobalt, would stand in a distinct class, appeared almost undoubted ; and it will be, I think, for the advantage of the inquiry, that I should consider them in a section apart by themselves. Further, if any other metals appeared to be magnetic, as these are, it would be right and expedient to include them in the same class.
2289. My first point, therefore, was to examine the metals for any indication of ordinary magnetism. Such an examination cannot be carried on by magnets any- thing short in power of those to be used in the further investigation ; and in proof of this point I found many specimens of the metals, which appeared to be perfectly free from magnetism when in the presence of a magnetic needle, or a strong horse- shoe magnet (2157.)5 that yet gave abundant indications when suspended near to one or both poles of the magnets described (2246.).
2290. My test, of magnetism was this. If a bar of the metal to be examined, about two inches long, was suspended (2249.) in the magnetic field, and being at first oblique to the axial line, was upon the supervention of the magnetic forces drawn into the axial position instead of being driven into the equatorial line, or remaining in some oblique direction, then I considered it magnetic. Or, if being near one magnetic pole, it was attracted by the pole, instead of being repelled, then I concluded it was magnetic. It is evident that the test is not strict, because, as before pointed out (2285.), a body may have a slight degree of magnetic force, and yet the power of the new property be so great as to neutralize or surpass it. In the first case, it might seem neither to have the one property nor the other ; in the second case, it might appear free from magnetism, and possessing the special property in a small degree.
2291. I obtained the following metals, so that when examined as above, they did not appear to be magnetic ; and in fact if magnetic, were so to an amount so small as not to destroy the results of the other force, or to stop the progress of the inquiry.
Antimony.
Bismuth.
Cadmium.
Copper.
Gold.
Lead.
Mercury.
Silver.
Tin.
Zinc.
2292. The following metals were, and are as yet to me, magnetic, and therefore companions of iron, nickel and cobalt : —
Platinum. Titanium.
Palladium.
2293. Whether all these metals are magnetic, in consequence of the presence of a
ACTION OF MAGNETS ON BISMUTH AND ANTIMONY. ,*H
little iron, nickel, or cobalt in them, or whether any of them are really so of them- selves, I do not undertake to decide at present ; nor do I mean to say that the metals of the former list are free. I have been much struck by the apparent freedom from iron of almost all the specimens of zinc, copper, antimony and bismuth, which I have examined ; and it appears to me very likely that some metals, as arsenic, &c, may have much power in quelling and suppressing' the magnetic properties of any portion of iron in them, whilst other metals, as silver or platinum, may have little or no power in this respect.
2294. Resuming the consideration of the influence excited by the magnetic force over those metals which are not magnetic after the manner of iron (2291.), I may state that there are two sets of effects produced which require to be carefully distin- guished. One of these depends upon induced magneto-electric currents, and shall be resumed hereafter (2309.). The other includes effects of the same nature as those produced with heavy glass and many other bodies (2276.).
2295. All the non-magnetic metals are subject to the magnetic power, and produce the same general effects as the large class of bodies already described. The force which they then manifest, they possess in different degrees. Antimony and bismuth show it well, and bismuth appears to be especially fitted for the purpose. It excels heavy glass, or borate of lead, and perhaps phosphorus; and a small bar or cylinder of it about two inches long, and from 0*25 to 05 of an inch in width, is as well fitted to show the various peculiar phenomena as anything I have yet submitted to exa- mination.
2296. To speak accurately, the bismuth bar which I employed was two inches long, 0*33 of an inch wide, and 0*2 of an inch thick. When this bar was suspended in the magnetic field, between the two poles, and subject to the magnetic force, it pointed freely in the equatorial direction, as the heavy glass did (2253.), and if disturbed from that position, returned freely to it. This latter point, though perfectly in accordance with the former phenomena, is in such striking contrast with the pheno- mena presented by copper and some other of the metals (2309.), as to require particular notice here.
2297- The comparative sensibility of bismuth causes several movements to take place under various circumstances, which being complicated in their nature, require careful analysis and explanation. The chief of these, with their causes, I will proceed to point out.
2298. If the cylinder electro- magnet (2246.) be placed vertically so as to present one pole upwards, that pole will exist in the upper end of an iron cylinder, having a flat horizontal face 2\ inches in diameter. A small indicating sphere (2266.) of bismuth hung over the centre of this face and close to it, does not move by the mag- netism. If the ball be carried outwards, halfway, for instance, between the centre and the edge, the magnetism makes it move inwards, or towards the axis (prolonged) of the iron cylinder. If carried still further outwards, it still moves inwards under
32 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
the influence of the magnetism, and such continues to be the case until it is placed just over the edge of the terminal face of the core, where it has no motion at all (here, by another arrangement of the experiment, it is known to tend in what is at present an upward direction from the core). If carried a little further outwards, the magnetism then makes the bismuth ball tend to go outwards or be repelled, and such continues to be the direction of the force in any further position, or down the side of the end of the core.
2299. In fact, the circular edge formed by the intersection of the end of the core with its sides, is virtually the apex of the magnetic pole, to a body placed like the bismuth ball close to it, and it is because the lines of magnetic force issuing from it diverge as it were, and weaken rapidly in all directions from it, that the ball also tends to pass in all directions either inwards or upwards, or outwards from it, and thus produces the motions described. These same effects do not in fact all occur when the ball, being taken to a greater distance from the iron, is placed in magnetic curves, having generally a simpler direction. In order to remove the effect of the edge, an iron cone was placed on the top of the core, converting the fiat end into a cone, and then the indicating ball was urged to move upwards, only when over the apex of the cone, and upward and outwards, as it was more or less on one side of it, being always repelled from the pole in that direction, which transferred it most rapidly from strong to weaker points of magnetic force.
2300. To return to the vertical fiat pole: when a horizontal bar of bismuth was suspended concentrically and close to the pole, it could take up a position in any direction relative to the axis of the pole, having at the same time a tendency to move upwards or be repelled from it. If its point of suspension was a little excentric, the bar gradually turned, until it was parallel to a line joining its point of suspension with the prolonged axis of the pole, and the centre of gravity moved inwards. When its point of suspension was just outside the edge of the flat circular terminating face, and the bar formed a certain angle with a radial line joining the axis of the core and the point of suspension, then the movements of the bar were uncertain and wavering. If the angle with the radial line were less than that above, the bar would move into parallelism with the radius and go inwards : if the angle were greater, the bar would move until perpendicular to the radial line and go outwards. If the centre of the bar were still further out than in the last case, or down by the side of the core, the bar would always place itself perpendicular to the radius and go outwards. All these complications of motion are easily resolved into their simple elementary origin, if re- ference be had to the character of the circular angle bounding the end of the core ; to the direction of the magnetic lines of force issuing from it and the other parts of the pole ; to the position of the different parts of the bar in these lines ; and the ruling principle that each particle tends to go by the nearest course from strong to weaker points of magnetic force.
2301 . The bismuth points well, and is well repelled (2296.) when immersed in water,
ACTION OF MAGNETS ON BISMUTH AND OTHER METALS. 33
alcohol, ether, oil, mercury, &c, and also when inclosed within vessels of earth, glass, copper, lead, &c. (2272.), or when screens of 0'75 or 1 inch in thickness of bismuth, copper, or lead intervene. Even when a bismuth cube (2266.) was placed in an iron vessel 2^ inches in diameter and 0' 17 of an inch in thickness, it was well and freely repelled by the magnetic pole.
2302. Whether the bismuth be in one piece or in very fine powder, appears to make no difference in the character or in the degree of its magnetic property (2283.).
2303. I made many experiments with masses and bars of bismuth suspended, or otherwise circumstanced, to ascertain whether two pieces had any mutual action on each other, either of attraction or repulsion, whilst jointly under the influence of the magnetic forces, but I could not find any indication of such mutual action : they ap- peared to be perfectly indifferent one to another, each tending only to go from stronger to weaker points of magnetic power.
2304. Bismuth, in very fine powder, was sprinkled upon paper, laid over the hori- zontal circular termination of the vertical pole (2246.). If the paper were tapped, the magnet not being excited, nothing particular occurred ; but if the magnetic power were on, then the powder retreated in both directions, inwards and outwards, from a circular line just over the edge of the core, leaving the circle clear, and at the same time showing the tendency of the particles of bismuth in all directions from that line (2299.J.
2305. When the pole was terminated by a cone (2246.) and the magnet not in ac- tion, paper with bismuth powder sprinkled over it being drawn over the point of the cone, gave no particular result ; but when the magnetism was on, such an operation cleared the powder from every point which came over the cone, so that a mark was traced or written out in clear lines running through the powder, and showing every place where the pole had passed.
2306. The bar of bismuth and a bar of antimony was found to set equatorially be- tween the poles of the ordinary horse-shoe magnet.
2307. The following list may serve to give an idea of the apparent order of some metals, as regards their power of producing these new effects, but I cannot be sure that they are perfectly free from the magnetic metals. In addition to that, there are certain other effects produced by the action of magnetism on metals (2309.) which greatly interfere with the results due to the present property.
Bismuth. Antimony. Zinc. Tin.
Cadmium. Mercury. Silver. Copper.
2308. I have a vague impression that the repulsion of bismuth by a magnet has been observed and published several years ago. If so, it will appear that what must mdcccxlvi. p
34 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
then have been considered as a peculiar and isolated effect, was the consequence of a general property, which is now shown to belong to all matter*.
2309. I now turn to the consideration of some peculiar phenomena which are pre- sented by copper and several of the metals when they are subjected to the action of magnetic forces, and which so tend to mask effects of the kind already described, that if not known to the inquirer they would lead to much confusion and doubt. These I will first describe as to their appearances, and then proceed to consider their origin.
2310. If instead of a bar of bismuth (2296.) a bar of copper of the same size be suspended between the poles (2247.), and magnetic power be developed whilst the bar is in a position oblique to the axial and equatorial lines, the experimenter will perceive the bar to be affected, but this will not be manifest by any tendency of the bar to go to the equatorial line; on the contrary, it will advance towards the axial position as if it were magnetic. It will not, however, continue its course until in that position, but, unlike any effect produced by magnetism, will stop short, and making no vibration beyond or about a given point, will remain there coming at once to a dead rest : and this it will do even though the bar by the effect of torsion or momentum was previously moving with a force that would have caused it to make several gyrations. This effect is in striking contrast with that which occurs when antimony, bismuth, heavy glass, or other such bodies are employed, and it is equally removed from an ordinary magnetic effect.
2311. The position which the bar has taken up it retains with a considerable de- gree of tenacity, provided the magnetic force be continued. If pushed out of it, it does not return into it, but takes up its new position in the same manner, and holds it with the same stiffness ; a push however, which would make the bar spin round several times if no magnetism were present, will now not move it through more than 20° or 30°. This is not the case with bismuth or heavy glass; they vibrate freely in the magnetic field, and always return to the equatorial position.
2312. The position taken up by the bar may be any position. The bar is moved a little at the instant of superinducing the magnetism, but allowing and providing for that, it may be finally fixed in any position required. Even when swinging with considerable power by torsion or momentum, it may be caught and retained in any place the experimenter wishes.
2313. There are two positions in which the bar may be placed at the beginning of
* M. de la Rive has this day referred me to the Bibliotheque Universelle for 1829, tome xl. p. 82, where it will he found that the experiment spoken of above is due to M. la Bajllif of Paris. M. la Baillif showed sixteen years ago that both bismuth and antimony repelled the magnetic needle. It is astonishing that such an experiment has remained so long without further results. I rejoice that I am able to insert this reference before the present series of these researches goes to press. Those who read my papers will see here, as on many other occasions, the results of a memory which becomes continually weaker ; I only hope that they will be excused, and that omissions and errors of that nature will be considered as involuntary. — M. F. December 30, 1845.
ACTION OF MAGNETS ON COPPER AND GOOD CONDUCTORS. 35
the experiment, from which the magnetism does not move it, the equatorial and the axial positions. When the bar is nearly midway between these, it is usually most strongly affected by the first action of the magnet, but the position of most effect varies with the form and dimensions of the magnetic poles and of the bar.
2314. If the centre of suspension of the bar be in the axial line, but near to one of the poles, these movements occur well, and are clear and distinct in their direction : if it be in the equatorial line, but on one side of the axial line, they are modified, but in a manner which will easily be understood hereafter.
2315. Having thus stated the effect of the supervention of the magnetic force, let us now remark what occurs at the moment of its cessation ; for during its continu- ance there is no change. If, then, after the magnetism has been sustained for two or three seconds, the electric current be stopped, there is instantly a strong action on the bar, which has the appearance of a revulsion (for the bar returns upon the course which it took for a moment when the electric contact was made), but with such force, that whereas the advance might be perhaps 15° or 20°, the revulsion will cause the bar occasionally to move through two or three revolutions.
2316. Heavy glass or bismuth presents no such phenomena as this.
2317- If, whilst the bar is revolving from revulsion the electric current at the magnet be renewed, the bar instantly stops with the former appearances and results (2310.), and then upon removing the magnetic force is affected again, and, of course, now in a contrary direction to the former revulsion.
2318. When the bar is caught by the magnetic force in the axial or equatorial position, there is no revulsion. When inclined to these positions there is ; and the places most powerful in this respect appear to be those most favourable to the first brief advance (2313.). If the bar be in a position at which strong revulsion would occur, and whilst the magnetism is continued be moved by hand into the equatorial or axial position, then on taking off the magnetic force there is no revulsion.
2319. If the continuance of the electric current and consequently of the magnetism be for a moment only, the revulsion is very little, and the shorter the continuance of the magnetic force the less is the revulsion. If the magnetic force be continued for two or three seconds and then interrupted and instantly renewed, the bar is loosened and caught again by the power before it sensibly changes its place ; and now it may be observed that it does not advance on the renewal of the force as it would have done had it been acted on by a first contact in that place (2310.) ; i. e. if the bar be in a certain place inclined to the axial position, the first supervention of the magnetic power causes it to advance towards the axial position ; but the bar being in the same place and the magnetic power suspended and instantly renewed, the second super- vention of force does not move the bar as the first did.
2320. When the copper bar is immersed in water, alcohol, or even mercury, the same effects take place as in the air, but the movements are, of course, not to the same extent.
f2
36 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
2321. When plates of copper or bismuth, an inch in thickness, intervene between the poles and the copper bar, the same results occur.
2322. If one magnetic pole only be employed the effects occur near it as well as before, provided that pole have a face large in proportion to the bar, as the end of the iron core (2246.) : but if the pole be pointed by the use of the conical termi- nation, or if the bar be opposite the edge of the end of the core, then they become greatly enfeebled or disappear altogether; and only the general fact of repulsion remains (2295.).
2323. The peculiar effects which have just been described are perhaps more strikingly shown if the bar of copper be suspended perpendicularly, and then hung opposite and near to the large face of a single magnetic pole, or the pole being placed vertically, as described (2246. 2263.), anywhere near to its side. The bar, it will be remem- bered, is two inches in length by 0'33 of an inch in width, and 0*2 of an inch in thick- ness, and as it now will revolve on an axis parallel to its length, the two smaller di- mensions are those which are free to move into new positions. In this case the establishment of the magnetic force causes the bar to turn a little in accordance with the effects before described, and the removal of the magnetic force causes a revulsion, which sends the bar spinning round on its axis several times. But at any moment the bar can again be caught and held in a position as before. The tendency on making contact at the battery is to place the longest moving dimension, i. e. the width of the bar, parallel to the line joining the centre of action of the magnet and the bar.
2324. The bar, as before (231 1.), is extremely sluggish and as if immersed in a dense fluid, as respects rotation on its own axis ; but this sluggishness does not affect the bar as a whole, for any pendulum vibration it has continues unaffected. It is very curious to see the bar, jointly vibrating from its point of suspension (2249.) and rotating on its axis, when first affected by the magnetic force, for instantly the latter motion ceases, but the former goes on with undiminished power.
2325. The same effect of sluggishness occurs with a cube or a globe of copper as with the bar, but the phenomena of the first turn and the revulsion cease (2310.2315.).
2326. The bars of bismuth and heavy glass present no appearance of this kind. The peculiar phenomena produced by copper are as distinct from the actions of these substances as they are from ordinary magnetic actions.
2327. Endeavouring to explain the cause of these effects, it appears to me that they depend upon the excellent conducting power of copper for electric currents, the gradual acquisition and loss of magnetic power by the iron core of the electro-mag- net, and the production of those induced currents of magneto-electricity which I de- scribed in the First Series of these Experimental Researches (55. 109.).
2328. The obstruction to motion on its own axis, when the bar is subjected to the magnetic forces, belongs equally to the form of a sphere or a cube. It belongs to these bodies, however, only when their axes of rotation are perpendicular or oblique
ACTION OF MAGNETS ON COPPER AND GOOD CONDUCTORS. 37
to the lines of magnetic force, and not when they are parallel to it; for the horizontal bar, or the vertical bar, or the cube or sphere, rotate with perfect facility when they are suspended above the vertical pole (2246.), the rotation and vibration being then equally free, and the same as the corresponding movements of bismuth or heavy glass. The obstruction is at a maximum when the axis of rotation is perpendi- cular to the lines of magnetic force, and when the bar or cube, &c. is near to the magnet.
2329. Without going much into the particular circumstances, I may say that the effect is fully explained by the electric currents induced in the copper mass. By reference to the Second Series of these Researches (160.)*, it will be seen that when a globe, subject to the action of lines of magnetic force, is revolving on an axis per- pendicular to these lines, an electric current runs round it in a plane parallel to the axis of rotation and to the magnetic lines, producing consequently a magnetic axis in the globe, at right angles to the magnetic curves of the inducing magnet. The magnetic poles of this axis therefore are in that direction which, in conjunction with the chief magnetic pole, tends to draw the globe back against the direction in which it is revolving. Thus, if a piece of copper be revolving before a north magnetic pole, so that the parts nearest the pole move towards the right-hand, then the right-hand side of that copper will have a south magnetic state, and the left-hand side a north magnetic state ; and these states will tend to counteract the motion of the copper towards the right-hand : or if it revolve in the contrary direction, then the right-hand side will have a south magnetic state, and the left-hand side a north magnetic state. Whichever way, therefore, the copper tends to revolve on its own axis, the instant it moves, a power is evolved in such a direction as tends to stop its motion and bring it to rest. Being at rest in reference to this direction of motion, then there is no residual or other effect which tends to disturb it, and it remains still.
2330. If the whole mass be moving parallel to itself, and be small in comparison with the face of the magnetic pole opposite to which it is placed, then, though it pass through the magnetic lines of force, and consequently have a tendency to the forma- tion of magneto-electric currents within it, yet as all parts move with equal velocity and in the same direction through similar magnetic lines of force, the tendency to the formation of a current is the same in every part, and there is no actual production of current, and consequently nothing occurs which can in any way interfere with its freedom of motion. Hence the reason that though the rotation of the bar or cube (2324. 2328.) upon its own axis is stopped, its vibration as a pendulum is not affected.
2331. That neither the one nor the other motion is affected when the bar or cube is over the vertical pole (2328.), is simply because in both cases (with the given dimensions of the pole and the moving metal) the lines of particles through which the induced currents tend to move are parallel throughout the whole mass ; and
* Philosophical Transactions, 1832, p. 168.
38 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
therefore, as there is no part by which the return of the current can be carried on, no current can be formed.
2332. Before proceeding to the explanation of the other phenomena, it will be necessary to point out the fact generally understood and acknowledged, I believe, that time is required for the development of magnetism in an iron core by a current of electricity; and also for its fall back again when the current is stopped. One effect of the gradual rise in power was referred to in the last series of these Researches (2170.). This time is probably longer with iron not well annealed than with very good and perfectly annealed iron. The last portions of magnetism which a given current can develope in a certain core of iron, are also apparently acquired more slowly than the first portions ; and these portions (or the condition of iron to which they are due) also appear to be lost more slowly than the other portions of the power. If electric contact be made for an instant only, the magnetism developed by the current disappears as instantly on the breaking of the current, as it appeared on its formation ; but if contact be continued for three or four seconds, breaking the contact is by no means accompanied by a disappearance of the magnetism with equal rapidity.
2333. In order to trace the peculiar effect of the copper, and its cause, let us con- sider the condition of the horizontal bar (2310. 2313.) when in the equatorial position, between the two magnetic poles, or before a single pole ; the point of suspension being in a line with the axis of the pole and its exciting wire helix. On sending an electric current through the helix, both it and the magnet it produces will conduce to the formation of currents in the copper bar in the contrary direction. This is shown from my former researches (26.), and may be proved, by placing a small or large wire helix-shaped (if it be desired) in the form of the bar, and carrying away the currents produced in it, by wires to a galvanometer at a distance. Such currents being produced in the copper, only continue whilst the magnetism of the core is rising and then cease (18.39.), but whilst they continue, they give a virtual magnetic polarity to that face of the copper bar which is opposite to a certain pole, the polarity being the same in kind as the pole it faces. Thus on the side of the bar facing the north pole of the magnet, a north polarity will be developed ; and on that side facing the south pole, a south polarity will be generated.
2334. It is easy to see that if the copper during this time were opposite only one pole, or being between two poles, were nearer to one than the other, this effect would cause its repulsion. Still, it cannot account for the whole amount of the repulsion observed alike with copper as with bismuth (2295.), because the currents are of but momentary duration, and the repulsion due to them would cease with them. They do, however, cause a brief repulsive effort, to which is chiefly due the first part of the peculiar effect.
2335. For if the copper bar, instead of being parallel to the face of the magnetic pole, and therefore at right angles to the resultant of magnetic force, be inclined,
ACTION OF MAGNETS ON COPPER AND GOOD CONDUCTORS. 39
forming, for instance, an angle of 45° with the face, then the induced currents will move generally in a plane corresponding more or less to that angle, nearly as they do in the examining helix (2333.), if it be inclined in the same manner. This throws the polar axis of the bar of copper on one side, so that the north polarity is not directly opposed to the north pole of the inducing magnet, and hence the action both of this and the other magnetic pole upon the two polarities of the copper will be to send it further round, or to place it edgeways to the poles, or with its breadth parallel to the magnetic resultant passing through it (2323.) : the bar therefore receives an impulse, and the angle of it nearest to the magnet appears to be pulled up towards the magnet. This action of course stops the instant the magnetism of the helix core ceases to rise, and then the motion due to this cause ceases, and the copper is simply subject to the action before described (2295.). At the same time that this twist or small portion of a turn round the point of suspension occurs, the centre of gravity of the whole mass is repelled, and thus I believe all the actions up to this condition of things is accounted for.
2336. Then comes the revulsion which occurs upon the cessation of the electric current, and the falling of the magnetism in the core. According to the law of mag- neto-electric induction, the disappearance of the magnetic force will induce brief currents in the copper bar (28.), but in the contrary direction to those induced in the first instance ; and therefore the virtual magnetic pole belonging to the copper for the moment, which is nearest the north end of the electro-magnet, will be a south pole ; and that which is furthest from the same pole of the magnet will be a north pole. Hence will arise an exertion of force on the bar tending to turn it round its centre of suspension in the contrary direction to that which occurred before, and hence the apparent revulsion ; for the angle nearest the magnetic pole will recede from it, the broad face (2323.) or length (2315.) of the bar will come round and face towards the magnet, and an action the reverse in every respect of the first action will take place, except that whereas the motion was then only a few degrees, now it may extend to two or three revolutions.
233/. The cause of this difference is very obvious. In the first instance, the bar of copper was moving under influences powerfully tending to retard and stop it (2329.) : in the second case these influences are gone, and the bar revolves freely with a force proportionate to the power exerted by the magnet upon the currents induced by its own action.
2338. Even when the copper is of such form as not to give the oblique resultant of magnetic action from the currents induced in it, when, for instance, it is a cube or a sphere, still the effect of the action described above is evident (2325.). When a plate of copper about three-fourths of an inch in thickness, and weighing two pounds, was sustained upon some loose blocks of wood and placed about O'l of an inch from the face of the magnetic pole, it was repelled and held off a certain distance upon the making and continuing of electric contact at the battery ; and when the battery cur-
40 DR. FARADAY'S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XX.)
rent was stopped, it returned towards the pole ; but the return was much more pow- erful than that due to gravity alone (as was ascertained by an experiment), the plate being at that moment actually attracted, as well as tending by gravitation towards the magnet, so that it gave a strong tap against it.
2339. Such is, I believe, the explanation of the peculiar phenomena presented by copper in the magnetic field ; and the reason why they appear with this metal and not with bismuth or heavy glass, is almost certainly to be found in its high electro- conducting power, which permits the formation of currents in it by inductive forces, that cannot produce the same in a corresponding degree in bismuth, and of course not at all in heavy glass.
2340. Any ordinary magnetism due to metals by virtue of their inherent power, or the presence of small portions of the magnetic metals in them, must oppose the de- velopment of the results I have been describing ; and hence metals not of absolute purity cannot be compared with each other in this respect. I have, nevertheless, observed the same phenomena in other metals ; and as far as regards the sluggishness of rotatory motion, traced it even into bismuth. The following are the metals which have presented the phenomena in a greater or smaller degree : —
Copper.
Silver.
Gold.
Zinc.
Cadmium.
Tin.
Mercury.
Platinum.
Palladium.
Lead.
Antimony.
Bismuth.
2341. The accordance of these phenomena with the beautiful discovery of Arago*, with the results of the experiments of Herschel and Babbage'|-, and with my own former inquiries (81.)^, are very evident. Whether the effect obtained by Ampere, with his copper cylinder and a helix §, was of this nature, I cannot judge, inasmuch as the circumstances of the experiment and the energy of the apparatus are not suffi- ciently stated ; but it probably may have been.
2342. As, because of other duties, three or four weeks may elapse before I shall be able to complete the verification of certain experiments and conclusions, I submit at once these results to the attention of the Royal Society, and will shortly embody the account of the action of magnets on magnetic metals, their action on gases and vapours, and the general considerations in another series of these Researches.
Royal Institution, Nov. 27, 1845.
* Annales de Chimie, xxvii. 363; xxviii. 325 ; xxxii. 213. I am very glad to refer here to the Comptes Rendus of June 9, 1845, where it appears that it was M. Arago who first obtained his peculiar results by the use of electro- as well as common magnets.
t Philosophical Transactions, 1825, p. 467. % Ibid. 1832, p. 146.
§ Bibliotheque Universelle, xxi. p. 48.