TRANSACTIONS OF THE SECTIONS. 19 
connected by 4 non-magnetic frame; yet such an arrangement would not experience 
any directional tendency, (since no one of the balls in it would experience either a 
resultant force or a resultant couple from the force of the field,) unless in virtue of 
changes in the states of magnetization of the balls induced by their mutual actions. 
Hence the mutual action of the parts of a row of balls, and, as is easily shown, of 
a row of cubes; or of a bar of any kind, must be taken into account before a true 
theory of their directional tendencies can be obtained. The author of this commu- 
nication, by elementary mechanical reasoning founded on what is known with cer- 
tainty regarding magnetic induction and magnetic action generally, shows that an 
elongated mass, in a uniform magnetic field, tends to place its length parallel to the 
lines of force, whetlier its inductive capacity be ferromagnetic or diamagnetic, pro- 
vided it be non-crystalline, because if ferromagnetic it becomes more, or if diamag- 
nétic, less intensely magnetized, if placed in such a position, than if placed with its 
length across the lines of force. But for all substances, whether ferromagnetic or 
diamagnetic, possessing so little capacity for induction as any of the known dia- 
magnetics, this tendency, depending as it does on the mutual action of the parts of 
the elongated mass, is, and probably will always remain, utterly imperceptible in 
experiment. All directional tendencies in bars of diamagnetic substance which have 
yet been, and probably all which can ever be discovered by experiment, are due 
either to some magnecrystallic property of their substances, or to the tendency of their 
ends or other moveable parts, from places of stronger towards places of weaker force, 
in varied magnetic fields, or to these two catisés combined, and in no respect to the 
inductive effects of the mutual influence of their parts. To consider the effects of a 
want of uniformity of the force, in a varied field, on the equilibrium of a ferromag- 
netic bar, the author quoted Faraday’s admirable statement of the law regarding the 
tendency of a ball or cube of diamagnetic substance, and referred to former papers, 
in which he had proved that, when applied to non-crystalline substances generally, 
with the proper modification for the case of ferromagnetics, it expresses with admi- 
rable simplicity the result of a mathematical investigation involving some of the 
most remarkable principles in the theory of attraction. From this it was shown, 
that if we conceive a ferromagnetic mass to be divided into very small cubes, each of 
these parts would; of itself, tend towards places of stronger force, and therefore 
that the bearing of the whole mass in a varied field will be produced partly by this 
tendency and partly by the tendency depending on the mutual inductive influence 
which alone exists when the field is uniform. The author then proceeded to illus- 
trate these theoretical views by a series of experiments. In some of them a steel bar 
magnet was used, and small soft iron wires, fixed in various positions on light wooden 
arms; were shown to be sometimes urged on the whole from places of stronger to © 
places of weaker force by their tendency to get into positions with their lengths along 
the lines of force. In others, a ring electro-magnet; consisting of insulated copper 
wire, rolled fifty times round as closely as possible to the circumference of a circle 
of the diameter stated, about 9 inches in diameter, fixed in a vertical plane at 
right angles to the magnetic meridian, was used, and a single cube of soft iron, 
placed in an excentric position on a long narrow pasteboard tray centrally sus- 
pended in the field of force by unspun silk, was attracted into the plane of the 
ring; but a row of three or four cubes placed touching one another in a line 
through the axis of stispension, settled as far from the plane as possible, in virtue 
of the tendency of an elongated mass to get its length along the lines of force. 
Two cubes placed in contact are found to be in stable equilibrium in the plane of the 
ring, or in oblique positions, or as far from the ring as possible, according to the 
greater or less distances at which they are placed in the tray, from the point of 
suspension. A number of equal and similar bars of a composition of wax and soft 
iron filings of different ferromagnetic strengths, suspended successively with their 
middle points in the centre of the magnet, settled in various positions. Those of 
them which were of greatest ferromagnetic capacity settled perpendicular to the 
plane of the ring or along the lines of force; others, with a smaller proportion of iron 
filings, had positions of stable equilibrium both in the plane of the ring and perpen- 
dicular to it; and others, with a still smaller proportion of iron filings, had their 
sole positions of stable equilibrium in the plane of the ring. The last-mentioned 
Ox 
