magnets depends also, to some extent, upon their form and
magnitude. It has been ascertained, that a bar magnet has the
best proportion when its thickness is about one-fourth and its
length twenty times its breadth.
1682. Horse-shoe magnets. Bar magnets are sometimes
shaped in the form of a horse-shoe, and are
hence called HORSE-SHOE MAGKETS, as repre-
sented in fig. 471. When magnets are con-
structed in this form, the distance between the
two poles ought not to be greater than the
thickness of the bar of which the magnet
consists. The surface of the steel forming
both bars, in horse-shoe magnets, should be
rendered as even and as well polished as pos-
1683. Methods of producing artificial mag-
nets by friction. Two methods of imparting
magnetism by friction are known as those of
DUHAMEL and JEriNUS. The former is sometimes called the
method of single touch, and the latter the method of double touch.
1684. Method of single touch. The method of Duhamel, or
of single touch, is practised as follows. The bar A' v',jig. 472.,
B L, A.
which is to be magnetized, is laid upon a block 'of wood L pro-
jecting at each end a couple of inches. Under the ends are
placed the opposite poles A and B of two powerful magnets, so
as to be in close contact with the bar to be magnetized. The
influence of the pole A will be to attract the boreal fluid of the
bar towards the end B', and to repel the austral fluid towards
the end A' ; and the effect of the pole B will be similar, that
is to say, to repel the boreal fluid towards the end B', and to
attract the austral towards the end A'. It is evident, therefore,
that if the coercive force of the magnetism of the bar A' B' be
not greater than the force of the magnets A and B, a decom-
position will take place by simple contact, and the bar A' B' will
be converted into a magnet, having its austral pole at A' and its
boreal pole at B' ; and, indeed, this will be accomplished even
though the coercive force of the bar A' B" be considerable, if it
be left a sufficient length of time under the influence of the
magnets A and B.
But without waiting for this, its magnetization may be accom-
plished immediately by the following process. Let two bar
magnets a and b be placed in contact with the bar A' B', to be mag-
netized near its middle point, but without touching each other,
and let them be inclined in opposite directions to the bar A! B',
at angles of about 30, as represented in the figure. Let the bar
which is applied on the side B' have its austral pule, and that
which is applied on the side A' its boreal pole in contact with the
bar A' B', and to prevent the contact of the two bars a and b, let
a small piece of wood, lead, copper, or other substance not sus-
ceptible of magnetism, be placed between them. Taking the
two bars a and b, one in the right and the other in left hand, let
them now be drawn in contrary directions, slowly and uniformly
along the bar A'B', from its middle to its extremities, and being
then raised from it, let them be again placed as before, near its
middle point, and drawn again uniformly and slowly to its ex-
tremities ; and let this process be repeated until the bar A' B'
has been magnetized.
It is evident that the action of the two magnetic poles a and
b will be to decompose the magnetic fluid of the bar A' B', and
that in this they are aided by the influence of the magnets A
and B, which enfeeble, as has been already shown, the coercive
This method is applicable with advantage to magnetize, in the
most complete and regular manner, compass needles, and bars
whose thickness does not exceed a quarter of an inch.
1685. Method of double touch. When the bars exceed this
thickness, this method is insufficient, and that of uEpinus, or the
method of double touch, is found more effectual. This method
is practised as follows.
The bars a and b are placed as before, but instead of being
held in the two hands are attached to a triangle, by which they
are maintained permanently in their position, and held together.
Being placed at the centre of the bar A'B', they are moved
together first to one extremity B', and then back along the length
of the entire bar to the other extremity A'. They are then
again drawn over the bar to B', and so backwards and for-
wards continuously until the bar is magnetized. The operation
is always terminated when the bars have passed over that half
of the bar A'B' opposite to that upon which the motion com-
menced. Thus if the operation commenced by moving the
united bars a b from the centre to the end B', it will be termi-
nated when they are moved from the extremity A' to the middle.
1686. Inapplicable to compass needles and long bars.
By this method a greater quantity of magnetism is developed
than in that of Duhamel, but it should never be employed for
magnetizing compass needles or bars intended for delicate ex-
periments, since it almost always produces magnets with poles
of unequal force, and frequently gives them consequent points
(1648), especially when the bars have considerable length.
1687. Magnetic saturation. Since the coercive force proper
to each body resists the recomposition of the magnetic fluids, it
follows that the quantity of magnetism which a bar or needle is
capable of retaining permanently, will be proportional to this
coercive force. If, by the continuance of the process of magne-
tization and the influence of very powerful magnets, a greater
development of magnetism be produced than corresponds with
the coercive force, the fluids will be recomposed by the mutual
attraction until the coercive force resists any further recom-
position. The tendency of the magnetic fluids to unite being
then in equilibrium with the coercive force, no further recom-
position will take place, and the bar will retain its magnetism
When the bar is in this state, it is said to be magnetized to
It has been generally supposed that when bars are surcharged
with magnetism they lose their surplus and fall suddenly to the
point of saturation, the recomposition of the fluids being in-
M. Pouillet, however, has shown that this recomposition is
gradual, and after magnetization there is even in some cases a
reaction of the fluids which is attended with an increase instead
of a diminution of magnetism. He observes that it happens not
unfrequently that the magnetism is not brought to permanent
equilibrium with the coercive force for several months.
1688. Limit of magnetic force. It must not be supposed
that by the continuance of the processes of magnetization which
have been described above, an indefinite development of mag-
netism can be produced. "When the resistance produced by the
coercive force to the decomposition of the fluids becomes equal
to the decomposing power of the magnetizing bars, all further
increase of magnetism will cease.
It is remarkable that if a bar which has been magnetized to
saturation by magnets of a certain power be afterwards sub-
mitted to the process of magnetization by magnets of inferior
power, it will lose the excess of its magnetism and fall to the
point of saturation corresponding to the magnets of inferior
1689. Influence of the temper of the bar on the coercive force.
Let a bar of steel tempered at a bright red heat be magnetized
to saturation, and let its magnetic intensity be ascertained by
the vibration of a needle submitted to its attraction. Let its
temper be then brought by annealing to that of a straw colour,
and being again magnetized to saturation, let its magnetic
intensity be ascertained. In like manner, let its magnetic in-
tensities at each temper from the highest to the lowest be
observed. It will be found that the bars which have the highest
temper have the greatest coercive force, and therefore admit of
the greatest development of magnetism, but even at the lowest
tempers they are still, when magnetized to saturation, susceptible
of a considerable magnetic force.
- Although highly tempered steel has this advantage of re-
ceiving magnetism of great intensity, it is, on the other hand,
subject to the inconvenience of extreme brittleness, and conse-
quent liability to fracture. A slight reduction of temper causes
but a small diminution in its charge of magnetism, and renders
it much less liable to fracture.
1690. Effects of terrestrial magnetism on bars. It has been
already shown that the inductive power of terrestrial magnetism
is capable of developing magnetism in iron bars, and, under
certain conditions, of either augmenting, diminishing, or even
obliterating the magnetic force of bars already magnetized. In
the preservation of artificial magnets, therefore, this influence
must be taken into account.
According to what has been explained, it appears that if
a magnetic bar be placed in the direction of the dipping-needle
in this hemisphere, the earth's magnetism will have a tendency
to attract the austral magnetism downwards, and to repel the
boreal upwards. If, therefore, the austral pole of the bar be
presented downwards, this tendency will preserve or even
augment the magnetic intensity of the bar. But if the magnet
be in the inverted position, having the boreal pole downwards,
opposite effects will ensue. The austral fluid being attracted
downwards, and the boreal driven upwards, a recombination of
the fluids will take place, which will be partial or complete ac-
cording to the coercive force of the bar. If the coercive force
of the bar exceed the influence of terrestrial magnetism, the
effect will be only to diminish the magnetic intensity of the
bar ; but if not, the effect will be the recomposition of the
magnetic force and the reduction of the bar to its natural
state ; but if the bar be still held in the same position, the
continued effect of the terrestrial magnet will be again to
decompose the natural magnetism of the bar, driving the austral
fluid downwards and repelling the boreal upwards, and thus
reproducing the magnetism of the bar with reversed polarity.
1691. Means of preserving magnetic bars from these effects
by armatures or keepers. It is evident, therefore, that when
it is desired to preserve magnetized bars unaltered, they must
be protected from these effects of the terrestrial magnet, and
the manner of accomplishing this is by means of ARMATURES or
When the magnetic bars to be preserved are straight bars of
equal length, they are laid parallel to each other, their ends
. corresponding, but with poles reversed,
^^| * so that the austral pole of each shall be
_|[ in juxtaposition with the boreal pole of
A the other, as represented \njig. 473.
A bar of soft iron called the keeper
is applied as represented at K, in contact with the two opposite
poles A and B', and another similar bar E 7 in contact with A' and
B, so as to complete the parallelogram. In this arrangement the
action of the poles A and B' upon the keeper K is to decompose
its magnetism, driving the austral fluid towards B' and the
boreal fluid towards A'. The boreal fluid of K exercises a
reciprocal attraction upon the austral fluid of A, and the austral
fluid of K exercises a corresponding attraction upon the boreal
fluid of B'. Like effects are produced by the keeper K' at the
opposite poles A' and B.
In this manner the decomposition of the fluids in the two
bars A B and A' B' is maintained by the action of the keepers
K and K'.
If the magnet have the horse-shoe form, this object is attained
by a single keeper, as represented \njig. 471.
The keeper K is usually formed with a round edge, so as to
touch the magnet only in a line, and not in a surface, as it
would do if its edge were flat. It results from experience that
a keeper kept in contact in this manner for a certain length of
time with a magnet, augments the attractive force, and appears
to feed, as it were, the magnetism.
1692. Magnetism may be preserved by terrestrial induction.
Magnetic needles, suspended freely so as to obey the attrac-
tion of terrestrial magnetism, do not admit of being thus pro-
tected by keepers ; but neither do they require it, for the austral
pole of the needle being always directed towards the boreal pole
of the earth, and the boreal pole of the needle towards the
austral pole of the earth, the terrestrial magnet itself plays the
part of the keeper, continually attracting each fluid towards its
proper pole of the magnet, and thus maintaining its magnetic in-
1693. Compound magnets. Compound magnets are formed
by the combination of several bar magnets of similar form and
equal magnitude, laid one upon another, their corresponding
poles being placed in juxtaposition.
A compound horse-shoe magnet, such as that represented in
Jig. 471., is formed in like manner of magnetized bars, superposed
on each other and similar in form, their corresponding poles
being placed in juxtaposition. These bars, whether straight or
in the horse-shoe form, are separately magnetized before being
combined by the methods already explained.
In the case of the horse-shoe magnet a ring is attached to the
keeper, and another to the top of the horse-shoe, /#. 471., so that
the magnet being suspended from a fixed point, weights may be
attached to the keeper tending to separate it from the magnet.
In this way horse-shoe magnets often support from ten to
twenty times their own weight.
1694. Magnetized tracings on a steel plate. If the pole of a
magnet be applied to a plate of steel of about one-tenth of an
inch thick and of any superficial magnitude, such as a square
foot, and be moved slowly upon it, tracing any proposed figure,
the line traced upon the steel plate will be rendered magnetic,
as will be indicated by sprinkling steel filings upon the plate.
They will adhere to those points over which the magnet has
been passed, and will assume the form of the figure traced upon
169.5. Influence of heat on magnetic bars. The influence of
heat upon magnetism, which was noticed at a very early period
in the progress of magnetic discovery, has lately been the subject
of a series of experimental researches by M. Kupffer, from which
it appears that a magnetic bar when raised to a red heat does
not lose its magnetism suddenly at that temperature, but parts
with it by slow degrees as its temperature is raised. This
curious fact was ascertained by testing the magnetism of the bar,
by the means explained in (1668), at different temperatures,
when it was found that at different degrees of heat it produced
different rates of oscillation of the test needle.
It was also ascertained that, in order to deprive a magnetic
bar of all its magnetism when raised to a given temperature, a
certain length of time was necessary.
Thus a magnetic bar plunged in boiling water, and retained
there for ten minutes, lost only a portion of its magnetism, and
after being withdrawn and again plunged in the water for some
length of time, it lost an additional portion of its attractive force ;
and by continuing in the same manner its immersion for the
same interval, its magnetic force was gradually diminished, a
part still, however, remaining after seven or eight such im-
A magnetic bar, when raised to a red heat, not only loses its
magnetism, but it becomes as incapable of receiving magnetism
from any of the usual processes of magnetization, as would be
any substance the most incapable of magnetism.
Astatic needle. All magnets freely suspended being subject to
the influence of terrestrial magnetism, the effects produced upon
them by other causes are necessarily compounded with those of
the earth. Thus, if a magnetic needle be exposed to the influence
of any physical agent, which, acting independently upon it, would
cause its north pole to be directed to the east, the pole, being at
the same time affected by the magnetism of the earth, which
acting alone upon it would cause it to be directed to the north,
will take the intermediate direction of the north-east. When,
in such cases, the exact effect of the earth's magnetism on the
direction of the needle is known, and the compound effect is
observed, the effect of the physical agent by which the needle is
disturbed may generally be eliminated and ascertained. It is,
nevertheless, often necessary to submit a magnetic needle to
experiments, which require that it should be rendered indepen-
dent of the directive influence of the earth's magnetism, and
expedients have accordingly been invented for accomplishing
A needle which is not affected by the earth's magnetism is
called an ASTATIC NEEDLE.
A magnetic needle freely suspended over a fixed bar magnet
will have a tendency, as already explained, to take such a
position that its magnetic axis shall be parallel to that of the
fixed magnet, the poles being reversed. Now if the fixed
magnet be placed with its magnetic axis coinciding with the
magnetic meridian, the poles being reversed with relation to
those of the earth, its directive influence on the needle will be
exactly contrary to that of the earth. "While the earth has a
tendency to turn the austral pole of the needle to the north, the
magnet has a tendency to turn it to the south. If these ten-
dencies be exactly equal, the needle will totally lose its polarity,
and will rest indifferently in any direction in which it may be
As the influence of the bar magnet on the needle increases
as its distance from it is diminished, and vice versa, it is evident
that it may always be placed at such a distance from it, that its
directive force shall be exactly equal to that of the earth.
In this case, the needle will be rendered astatic.
A needle may also be rendered astatic by connecting with it
a second needle, having its magnetic axis parallel and its poles
reversed, both needles having equal magnetic forces. The com-
pound needle thus formed being freely suspended, the directive
power of the earth on the one will be equal and contrary to its
directive power on the other, and it will consequently rest in-
definitely in any direction.
It is in general, however, almost impracticable to ensure the
exact equality of the magnetism of two needles thus combined.
If one exceed the other, as is generally the case, the compound
will obey a feeble directive force equal to the difference of their
BOOK THE THIRD.
ELECTRICAL ATTRACTIONS AND REPULSIONS.
1696. Electrical effects. If a glass tube being well dried be
briskly rubbed with a dry woollen cloth, the following effects may
The tube, being presented to certain light substances, such as
feathers, metallic leaf, bits of light paper, filings of cork or pith
of elder, will attract them.
If the friction take place in the dark, a bluish light will be
seen to follow the motions of the cloth.
If the glass be presented to a metallic body, or to the knuckle
of the finger, a luminous spark accompanied by a sharp cracking
sound, will pass between the glass and the finger.
On bringing the glass near the skin, a sensation will be pro-
duced like that which is felt when we touch a cobweb.
The same effects will be produced by the cloth with which
the glass is rubbed as by the glass itself.
An extensive class of bodies submitted to the same kind of
mutual friction produce similar effects.
1697. Origin of the name Electricity. The physical agency
from which these and like phenomena arise has been called
ELECTRICITY, from the Greek word j/Xe/crpov (electron), signifying
amber, that substance having been the first in which the pro-
perty was observed by the ancients.
To study the laws which govern electrical forces, let an ap-
paratus be provided, called an electric pendulum, consisting of
a small ball B', fig. 474., about the tenth of an inch in diameter,
turned from the pith of elder, and suspended, as represented in
the figure, by a fine silken thread attached to a convenient
If the glass tube, after being rubbed as above described, be
brought into contact successively with two pith balls thus
suspended, and then separated from them, a property will be
imparted to the balls in virtue of which they will be repelled
by the glass tube when it is brought nearer them, and they will
in like manner repel each other when brought into proximity.
Thus, if the glass tube s, fig. 474., be brought nearer the ball
B', the ball will depart from its vertical position, and will incline
itself from the tube in the position B.
If the two balls, being previously brought into contact with
the tube, be placed near each other, as \\\fig. 475., they will in-
cline from each other, departing from the vertical positions B
and B', and taking the positions b and b\
1698. The electric fluid. These effects are explained by
the supposition that a subtle and imponderable fluid has been
developed upon the glass tube which is self-repulsive ; that by
touching the balls, a portion of this fluid has been imparted to
them, which is diffused over their surface, and which, for reasons
that will hereafter appear, cannot escape by the thread of sus-
pension ; that the fluid remaining on the glass tube repels this
fluid diffused on the balls, and therefore repels the balls them-
selves which are invested by the fluid; and, in fine, that the fluid
diffused on the one ball repels and is repelled by the fluid dif-
fused on the other ball, and that the balls being covered by the
fluid are reciprocally repelled.
A vast body of phenomena, the most important of which
will be described in the following chapters, have converted this
supposition into a certainty, accepted by all scientific authorities.
ELECTRICAL ATTRACTIONS AXD REPULSIONS. 191
The fluid producing these effects is called the ELECTRIC
1699. Positive and negative electricity. If the hand which
holds the cloth be covered with a dry silk glove, the cloth, after
the friction with the glass, will exhibit the same effects as above
described. If it be brought into contact with the balls and se-
parated from them, it will repel them, and the balls themselves
will repel each other.
It appears, therefore, that by the friction the electric fluid is at
the same time developed on the glass and on the cloth.
If after friction the glass be brought into contact with one
ball Bjjtfy. 475., and the cloth with the other B', other effects
will be observed. The glass, when presented to the ball B', will
attract it, and the cloth presented to the ball B will attract it.
The balls, when brought near each other, will now exhibit
mutual attraction instead of repulsion.
It follows, therefore, that the electric fluid developed by friction
on the cloth differs from that developed on the glass, inasmuch
as instead of being characterized by reciprocal repulsion they
are mutually attractive.
The supposition, therefore, which has been briefly stated above
will require modification.
1700. Hypothesis of a single electric fluid. According to
some, the effect of the friction is to deprive the cloth of a portion
of its natural charge of electricity, and to surcharge the glass
with what the cloth loses ; and accordingly the glass is said to be
positively, and the cloth negatively electrified.
On this supposition, bodies in their natural state have always
a certain charge or dose of the electric fluid, the repulsive effect
of which is neutralized by the attraction exercised by the body
upon it. The electric equilibrium which constitutes this natural
state may be deranged, either by overcharging the body with
the electric fluid, or by withdrawing from it a part of what it
naturally possesses. In the former case, the repulsion of the
surplus charge not being neutralized by the attraction of the
body takes effect. In the latter case, the attraction of the body
being more than equal to the repulsion of the charge of electri-
city upon it, will take effect upon any electricity which may
come within the sphere of its action.
This, which is called the SINGLE FLUID THEORY, was the
hypothesis adopted by FRANKLIN, and after him by most English
electricians until recently, when phenomena were developed
in experimental researches, of which it failed to afford a satis-
factory explanation ; and, accordingly, the hypothesis of two
fluids, which was generally received on the Continent, has