Two hours.
The clock is the exam.

For most of the eighteenth century a ship's captain could fix his latitude within minutes of coming on deck, yet could only guess at his longitude. Latitude follows from the height of the sun at noon, a measurement any competent officer could take with a quadrant. Longitude, by contrast, is a question about time: because the earth turns fifteen degrees each hour, a navigator who knows the difference between local noon and the time at a reference port can convert that gap directly into a distance east or west. The obstacle was never the arithmetic. It was that no clock could keep reference time aboard a rolling, salt-soaked, temperature-swinging ship.

The British Board of Longitude, established in 1714, offered a prize for a solution accurate to half a degree after a voyage to the West Indies. Most of the astronomers who advised the Board assumed the answer would come from the sky. Their preferred method, lunar distances, required the observer to measure the angle between the moon and a fixed star and then work through several pages of tables and corrections. It was ingenious and it was free of moving parts, but a single computation could occupy four hours, and it failed whenever cloud closed in.

John Harrison, a joiner from Lincolnshire with no formal training in horology, took the mechanical route the Board had quietly dismissed. His fourth timekeeper, a watch some thirteen centimetres across, lost only a handful of seconds on an Atlantic crossing in 1761. The Board did not pay him in full for more than a decade, and it is tempting to read that delay as simple prejudice against a provincial craftsman. The likelier explanation is institutional. The Board had been asked to certify a method, not a single object, and one watch that worked told it nothing about whether a hundred could be built to the same standard.