38 
SUPPLEMENT. 
to all situations, and particularly when it has now been put to 
the test in Baffin’s Bay. 
The memoir, written by Captain Flinders on this subject, is 
recorded in the Philosophical Transactions of the Royal Society, 
for the year 1805, from this it appears to have been his opinion, 
that the error of variation consequent on a change in the direction 
of the ship’s head, was produced by the combined force of terres¬ 
trial magnetism and ferruginous attraction within the ship. 
In the year 1812, the Lords Commissioners of the Admiralty 
ordered experiments to be made on board of five different vessels 
in the king's ports, “ with the view of ascertaining the particular 
causes of error to which Captain Flinders had adverted, or of 
obtaining some general results from an inquiry so intimately 
connected as it appeared to be with the improvement of naviga¬ 
tion.” These experiments, as far as they went, tended to estab¬ 
lish the fact, and to justify the opinion of Captain Flinders. Still, 
however, more information was wanted respecting the subject, 
for the purpose of discovering a rule, that would enable observers 
to find the true quantity of error in any place and under all 
circumstances. 
Although the experiments above mentioned gave some insight 
into the causes of this variation, they were insufficient to explain 
them perfectly; nor is it probable that we shall soon be made 
acquainted with them, ignorant as we are of the nature of many 
physical appearances of familiar occurrence. Though it wou.d 
perhaps be possible, in the present highly improved state of na¬ 
vigation, for one thoroughly versed in seamanship and nautical 
astronomy to conduct a ship in safety from England to any port 
in the world, without the aid of the mariner’s compass, yet, in 
cloudy and tempestuous weather, or in confined waters and sur¬ 
rounded by land, his doubt and anxiety could only be relieved, 
or confidence given to his mind, by the compass. It is therefore 
necessary that this instrument should be rendered as unerring a 
guide as possible, and this can only be done by a certain, uni¬ 
versal, and invariable mode of finding the true variation, at all 
times and places, and under all circumstances. 
