IN WOOD-BUILT AND IRON-BUILT SHIPS. 
55 
azimuth differing 90° from that of the disturbing mass. And, if it be found that 
when a ship’s head is in the quadrant between N. and E., or between S. and W., the 
needle deviates to the right, and the opposite way for the remaining quadrants ; or 
that, in respect to the quadrants of azimuth of ship’s head, the quadrantal deviation 
follows the law H 1 — which I shall call “positive quadrantal deviation;” the 
inference is that the deviation is of the same kind as would be produced by a mass 
of iron at the same level as the compass, either headward or sternward of the com- 
pass : and it may be neutralized by placing a mass of iron at the same level as the 
compass, either on the starboard or on the port side. But if the deviation follow the 
law — I h in respect of the four quadrants of azimuth, which I shall call “ nega- 
tive quadrantal deviation,” it may be neutralized by placing a mass of iron at the 
same level as the compass, either headward or sternward of the compass. All these 
laws I have abundantly confirmed by experiment. 
These being premised, the laws of the deviation of the compass produced by the 
transient induced magnetism of a ship, as shown by swinging the ship round in 
a given locality, will (according to the theory to which 1 have referred) be as fol- 
lows ; — 
(1) . There will be a force, similar to the force of a polar-magnet, and producing 
a polar-raagnet-deviation. In northern magnetic latitudes, the nature of the effect 
will usually be the same as if the boreal magnetism were towards the ship’s head : in 
southern magnetic latitudes, it will be usually the same as if the austral magnetism 
were towards the ship’s head. The absolute magnitude of the polar-magnet-force 
will be a multiple of the vertical terrestrial magnetism ; the proportion which it 
bears to the terrestrial directive force, which is the proportion of AB to AC in fig. 1 
(supposing no other polar-magnet-force to act), will be a multiple of the tangent of 
dip, without regard to the absolute force. 
(2) . There will be a quadrantal deviation ; and this deviation will be the same in 
all magnetic latitudes, and whatever be the magnitude of the earth’s magnetic force. 
It will usually be a positive quadrantal deviation. 
These are the disturbances that are produced by transient induced magnetism only. 
But if the iron that enters into the composition of a ship possess independent polar 
magnetism similar to that of a magnetized steel bar {i. e. not depending on the ter- 
restrial magnetism at the present moment for its existence ; and not changing its 
amount or quality or direction in regard to the ship’s keel, while the ship is swung 
round in different positions), which from the slowness of its changes, though pro- 
bably more variable than that of a steel bar, I propose to call “subpermanent mag- 
netism;” it will be necessary for us to consider how the expression for the effects 
of this subpermanent magnetism can be most easily combined with those for the 
induced transient magnetism. It is readily seen that the polar-magnet-force of sub- 
permanent magnetism must be combined with the polar-magnet-force of induced 
transient magnetism ; and that, at a given locality, they cannot be separated. In 
