Astronomical and Nautical Collections. 321 
pressure to overcome the pressure of the air which confines the fluid 
to the surface. This fact is universally true, and is independent of 
the form or magnitude of the pieces of iron which are in contact, 
and of the degree of their magnetism, or of the force of the magnet 
which influences them: however intimate the contact may be, and 
however long it may have lasted, the fluid never passes from any 
one piece into another; whence it is natural to infer, that no sen- 
sible quantity of magnetism is ever transported from one part to 
another of the same piece of iron; and that the boreal and austral 
fluids, contained by the metal in its natural state, undergo insensible 
displacements only within it, when they are separated from one 
another by any exterior action. This conclusion is also equally 
applicable to those magnetized bodies, which retain the magnetism, 
that has been excited in them, either by the continued influence of 
a strong magnet, or by other means: the orly distinction between 
these substances and soft iron is, that there exists in them a force 
peculiar to each substance, known by the name of the coercive 
force, of which the effect is, to arrest the particles of both fluids in 
the situations which they occupy, and to oppose in this manner 
first the separation of the two fluids, and then their return to their 
natural union. 
A question occurs, relating to this subject, which does not ap- 
pear to have hitherto excited the attention of natural philosophers, 
which arises however very naturally from considering the magnetic 
- fluid as always belonging to the same constituent particles of the 
magnetized bodies. It is not only not demonstrated that this fluid 
is identical with the electric fluid, but it is not even necessary to 
suppose that the phenomena of magnetism are produced in all 
bodies by a fluid possessing every where the same intensity of at- 
‘tractive or repulsive action, and therefore requiring to be consi- 
dered as the same fluid in different substances. The identity of 
the electric fluid is shown by its passing from one conducting body 
into another, in such a manner as to preserve all its properties, and 
to exercise, in the same circumstances, the same attractions or 
repulsions ; but no such test as this can be applied in the case of 
magnetism, and we cannot decide from mere reasoning, whether 
