How sailors found longitude before satellites: a race against the clock

Latitude is written in the sky

Finding how far north or south you are has always been fairly easy. At night in the northern hemisphere, the height of the Pole Star above the horizon is roughly your latitude. By day, you measure the Sun's height at noon and look up a table for the date. Sailors were doing this with quadrants and backstaffs long before 1700, and a careful navigator could usually get within a few tens of kilometres.

East and west are different. The sky over a ship at 30° west looks exactly like the sky over a ship at 40° west. It just shows up later. The Earth turns 360° in 24 hours, which is 15° an hour, or one degree every four minutes. So longitude is really a difference in time. If you know it is noon where you are (the Sun is at its highest) and you know what time it is at home, the gap tells you how far round the world you have come.

The catch is the second half. To know the time at home, you have to carry it with you. In 1700 no clock could do that on a rolling, damp ship whose temperature swung from the Channel to the tropics. Pendulum clocks were useless at sea, and the best watches could drift by minutes a day. Every four minutes of drift put you a degree out: up to 111 km at the equator.

Four ships on the rocks

On the night of 22 October 1707, a British fleet returning from the Mediterranean under Admiral Sir Cloudesley Shovell ran onto the rocks of the Isles of Scilly, off the tip of Cornwall. Four warships sank, including the flagship HMS Association. Somewhere between 1,400 and 2,000 sailors drowned, one of the worst disasters in British naval history.

The wreck became the classic story of the longitude problem, though the real cause was messier. Later studies point to latitude errors, charts that put Scilly several miles out of place, a northward current and bad compasses. Still, the fleet did not know where it was, and people at the time blamed the missing longitude. Seven years later Parliament acted.

A prize for the impossible

The Longitude Act of 1714 offered £10,000 for a method good to within one degree (60 nautical miles), £15,000 within 40 miles and £20,000 within 30 miles, tested on a voyage from Britain to the West Indies. It was a fortune, and it attracted cranks as well as astronomers.

The serious answer came from John Harrison, a self-taught carpenter from Lincolnshire. His first sea clock, H1, sailed to Lisbon in 1736 and did well enough to win development money. He spent the next two decades on H2 and H3 before switching to something radical: a large watch, about 13 cm across, finished in 1759. It is now known as H4.

In November 1761 H4 left Portsmouth for Jamaica aboard HMS Deptford, looked after by Harrison's son William. After 81 days at sea it was about 5 seconds out. That is roughly a nautical mile of longitude. The Board of Longitude was not satisfied and asked for a second trial. In 1764, on the voyage to Barbados, H4 was 39 seconds out, still an error of under 10 miles.

Harrison was paid £10,000 in 1765, but only on condition that he explained the design and that copies could be made. The Board never handed over the rest of the prize. In 1773, after King George III took his side, Parliament granted him another £8,750. By then he was 80.

Try it: how bad is a bad clock?

Set how many seconds a day the ship's clock gains or loses, how long you have been at sea, and your latitude. The strip shows where you think you are (blue) and where you really are (red), with the Act's three prize zones drawn around the truth.

Things to try:

• Start with a watch that drifts a minute a day. After a six-week crossing you are over 1,000 km out: the width of a sea.
• Find the drift that just wins £20,000 on that voyage. It is only a few seconds a day.
• Move the latitude towards 60°. The same clock error costs half as many kilometres, because lines of longitude bunch together towards the poles.

The rival: reading time off the Moon

Astronomers had a different plan. The Moon moves across the stars by about its own width every hour, so it acts like the hand of a clock in the sky. Measure the angle between the Moon and a bright star with a sextant, correct it, and compare it with tables that predict that angle for every few hours of Greenwich time. The match tells you the time at Greenwich.

This lunar distance method only became workable once the German astronomer Tobias Mayer produced accurate lunar tables. Nevil Maskelyne, later Astronomer Royal, turned them into the Nautical Almanac, first issued for 1767, and Mayer's widow was awarded £3,000. It needed no expensive clock, just a sextant, a clear sky and a navigator who could handle a long calculation. At best it gave longitude to about a quarter of a degree.

For decades the two methods ran side by side, the lunar distance to check the clock. Chronometers were expensive at first. But Larcum Kendall's copy of H4 went round the world with James Cook, and as chronometers got cheaper they won. Radio time signals in the 20th century and satellites later made the question routine. The core idea never changed: finding your place east or west means knowing two times at once.