The magnetic properties of magnetite, the lodestone used in compasses, have been known for thousands of years, but it was not until the 12th century that Chinese navigators began using the mineral in compasses to help sailors find magnetic north. Early compasses consisted of a magnetized bar or needle floating in a dish of water. More advanced instruments placed the needle on a pivot so it hovered above a card bearing a compass rose, whose points correlated to the Temple of the Winds devised by the ancient Greeks.
Compasses were popularized in the 15th century by Portuguese sailors, whose instruments sometimes featured magnetized cards that pivoted rather than needles. Other compasses were designed to be read from below, say by a sailor laying in his bunk. By the 16th century, the world’s greatest European compass makers were based in Nuremberg and Bruges.
As navigation aids, compasses are certainly better than nothing, but they are not without their shortcomings. The first problem is that magnetic north and true north are not the same, which means compasses had to be designed to compensate for the resulting magnetic declination. The needles in compasses were also affected by the steel in ships, which was offset in the early 1800s by Commander Matthew Flinders, who devised his Flinders bars.
And then there is the sea itself, which practically guarantees that a compasses will never be read at rest and perfectly horizontal. To solve this problem, gimbals were devised to keep the compass constantly suspended; the compass was typically placed within a binnacle which housed the gimbal and supported other instruments. By the 20th century, this technology was improved upon and adapted to automobiles, which were outfitted with compasses to help drivers find their way on unmarked roads.