178 
LINES OF MAGNETIC DECLINATION IN THE ATLANTIC. 
artificial counteractions were employed, the magnetism of the ship had been regarded 
either as wholly induced by the magnetic action of the earth and as varying simul- 
taneously with variations in the inducing cause, or as partly due to induction, and 
partly to the permanent magnetism of certain portions of the ship’s iron ; in which 
latter case the part of the disturbance occasioned by induction was considered to be 
variable in the manner already described ; and the part occasioned by the permanent 
magnetism to be constant, or nearly so. 
Upon the hypothesis of the whole disturbance being occasioned by induced mag- 
netism, its amount in any particular direction of the ship’s head should be the same, 
or nearly so, in north and south dips of equal amount, but should have opposite signs; 
that is to say, the disturbance which was towards the west when the north end of the 
needle dipped, should be towards the east when the south end of the needle dipped, 
and should be to the same amount : and if the further assumption were correct, that 
the induced magnetism of the ship changed simultaneously with changes in the 
terrestrial dip, as is known to be the case with soft iron, the disturbance might be 
altogether (at least approximately) prevented, by a counteracting mass of soft iron 
disposed suitably in reference to the place of the compass and to the resultant of the 
ship’s magnetic action. 
On the supposition that the disturbances were due partly to the induced magnetism 
of certain portions of a ship’s iron, and partly to the permanent magnetism of other 
portions, the calculation of corrections would become more complex, as terms must 
be introduced to represent both a variable and a constant effect ; and counteraction 
by means of soft iron would no longer meet the case. But a combination of perma- 
nent magnets and of soft iron, each suitably disposed, might, as was supposed, accom- 
plish and preserve an approximate compensation, if the magnets and the permanently 
magnetic portion of the ship’s iron maintained their magnetic relations unaltered, 
and if the changes of the induced magnetism of the ship were as simultaneous with 
changes in the terrestrial magnetism as they were presumed to be in soft iron. 
The observations which were made on the disturbance of the compass needle of 
the Erebus and Terror in the river Thames, where the magnetic dip was about 69° 
7iorth, and at Hobarton, where the dip was between 70° and 71° south (the ships in 
both cases having remained several months in the localities of the respective dips), 
showed that in the interval between the two sets of observations a change had taken 
place in the disturbance, corresponding in kind, and almost precisely in degree, with 
the hypothesis of induced magnetism. The disturbance was in the opposite direction 
at Hobarton to what it had been in the Thames : in the one case the north pole of 
the compass needle was drawn towards the fore part of the ship, and in the other 
case the south pole. The amount of disturbance in the one direction in the Thames, 
and in the opposite direction at Hobarton, was so nearly the same — the terrestrial 
dip having also nearly the same numerical value at the two stations, but with oppo- 
site signs, — as fully to bear out the inference, that in those two ships the chief part 
