LINES OF MAGNETIC DECLINATION IN THE ATLANTIC. 
179 
if not the whole of the disturbance of the compass was occasioned by induced mag- 
netism. 
Such being the case, it might naturally have been expected that formulae founded 
on M. Poisson’s investigations regarding the induced magnetism of ships, which re- 
presented so well the change that had taken place between England and Van Diemen 
Island, would also represent the disturbance which had been found to take place at 
stations visited by the ships in the intermediate passage ; and that the i-esult of azi- 
muths observed in the same geographical position of the ship, with her head on 
different points of the compass, would be brought into agreement with each other, at 
any period of the voyage, by corrections computed by the formulte of which the 
variable coeflScients were taken as varying with the changes of the terrestrial dip. 
Such however was by no means the case. A table of corrections was computed by 
the appropriate formulae for each of the thirty-two points and for every degree of 
north and south dip ; the values of the coefficients in the formulae being derived from 
the observations in the Thames and at Hobarton ; and those which were variable 
being assumed, in conformity with the hypothesis, to vary according to the dip. On 
comparing this table with the observations at intermediate times and stations, it was 
immediately perceived that in order to suit the table to the observations, it was 
necessary to enter the table, not with the dip at the time and place of the observa- 
tion to be corrected, but with a dip which had been passed through by the ship several 
days antecedently ; and on a more close and general examination, this was found to 
be the systematic and consistent result of the whole comparison. 
This result by no means contradicts the inference previously drawn, and based on 
the observations in the Thames and at Hobarton, viz. that the disturbances in the 
Erebus and Terror were chiefly if not wholly ascribable to induced magnetism ; for 
it is quite conceivable that portions of a ship’s iron, which are not permanently 
magnetic on the one hand, nor perfectly soft so as to undergo instantaneous change 
with changes of the dip on the other hand, may still derive magnetism by induction 
from the earth, which may conform gradually veLlhev than instantaneously to the changes 
of terrestrial magnetism corresponding to changes of the ship’s place ; so that after 
an interval of greater or less duration, the variation of the magnetic state which is 
characteristic of induced magnetism may be as complete in such portions of the iron 
as in those in which the change takes place instantaneously: but it is inconsistent 
with the proposed counteraction of the induced portion of the disturbance by means 
of soft iron, unless a degree of retentive force could be given to the soft iron which 
should be precisely equivalent to that of the general resultant of all the iron in a ship 
which is not permanently magnetic ; and which doubtless varies considerably in 
different ships. 
Since 1843, when the fifth number of the Magnetic contributions was printed, I 
have examined the observations made in several ships which have passed from one 
hemisphere to the other, and have found them, without a single exception, cor- 
