Terrestrial Magnetism, 129 
billed action of the ship’s iron was exerted, and at the 
exact distance from the compass at which its effect would 
be equal to that of the ship’s iron: the amount of deviation 
was consequently doubled; and this would be the case 
under all circumstances in any part of the world. This 
point being ascertained, the ball was to be laid aside 
until the amount of the deviation of the compass in any 
other latitude was sought, and then to be replaced. The 
angle which the present direction of the needle makes 
with that which it had previous to the application of the 
ball, shows the amount of the actual deviation produced 
by the iron of the ship. Strictly speaking, it is not the 
angle of deviation, but the tangent of that angle ; but 
as in small angles the tangents are very nearly in the 
ratio of their arcs, it may be taken as correct without 
any sensible error. 
The truth of the above practical rule was ascertained 
by experiments on a very extended scale, on considerable 
masses of iron, such as balls and shells of all denomina¬ 
tions, which from their forms were the most favourable 
to the application of mathematical investigation. It was 
proved by repeated experiments, that in all spheres of 
iron there was a “ plane of neutrality” passing through 
their centres ; i. e. when a magnetised needle is placed in 
any part of a plane passing through the centre of a ball 
of iron, and at right angles to the direction of the dipping 
needle in that place, that ball produces no disturbance. 
This plane is therefore inclined to the horizon at an angle 
equal to the complement of the angle of the dip at that 
place. Another plane of neutrality was also found to exist 
in a vertical plane running through the centre of a ball, 
and another hollow sphere which might be supposed to 
extend around it, and concentric with it, and in the plane 
of the magnetic meridian. It was from these investiga¬ 
tions that a mathematical law was deduced to show the 
vol. i. xo. 11. k. 
