8 
GREENLAND VOYAGE. 
towards the south, retarded;—in the same way 
that the vibrations of a compass-needle are acce¬ 
lerated or retarded, when near a magnet, accord¬ 
ingly as dissimilar or similar poles are presented 
to each other. The changes, in the rates of chro¬ 
nometers, arising from this cause, are very various, 
depending not only on the degree of magnetic 
energy of the balances, but also very materially on 
the positions of the instruments when their land 
rate was determined. It will sometimes happen, 
that the rate is taken in a position where the mag¬ 
netic axis of the balance coincides pretty nearly 
with that of the earth: the error, in such a case, 
will be a maximum ; because, at sea, the position 
is so varied by changes of course, as to have a 
tendency to produce a mean action, and a mean 
rate. All chronometers, therefore, when rated on 
shore, should be tried in at least four different po¬ 
sitions, such as with the 12 o’clock mark present¬ 
ed to the four cardinal points. This would evi¬ 
dently afford a mean rate that would greatly en¬ 
hance the value of the instrument, by giving it 
the best chance in the event of a change in its 
going. 
With the assistance of Mr Thomas Davies, 
of the firm of Litherland and Davies of Li¬ 
verpool, I have made a number of experiments 
on the change of rate in chronometers by posi- 
