The great circles of longitude that sweep through the poles of a classroom globe seem fixed and solid. But imagine trying to locate them from the deck of a ship bobbing in a vast ocean! John Harrison, a self-taught mechanical genius who pioneered a failsafe method of establishing longitude at sea, earned the gratitude of every sailor who has ever wondered what dangers lurk beyond the horizon. But, to do it, he had to take on the British scientific establishment in a battle that took a lifetime to win.

The cruel sea
A life on the ocean wave was not for the faint-hearted in the early 18th century. With no reliable way of knowing how far east or west you had sailed, seamen were constantly losing their way or foundering on rocks that took them by surprise. Determining longitude was the greatest technical challenge of the age. In 1714, that challenge was formalised by an act of parliament offering £20,000 to anyone who could determine longitude at sea to within 30 nautical miles.

Staring at the sun
Longitude and time are locked together. Fifteen degrees of longitude is equal to one hour's time difference. If you compare local time with the time at a fixed reference point, you can work out how far east or west of that reference point you are. An 18th-century mariner could determine local time by setting his watch according to the sun. But keeping an accurate record of the reference time was impossible. Pendulums failed aboard a pitching ship and changes of temperature and humidity during a sea voyage interfered with all known timepieces. Instead, astronomers insisted that the only way to calculate longitude was by the lunar method - painstaking observation of the moon's position in relation to the sun and other celestial bodies.

A simple man
The man who tamed longitude was not a traditional academic. John Harrison was born in 1693, the son of a Yorkshire joiner. John had a precocious talent for mechanics and, by 1727, he was designing a maritime clock more accurate than anything yet seen. In 1736, his clock, H1, underwent successful sea trials to Lisbon. As a result, Harrison appeared before the Board of Longitude, a body of academics appointed to oversee any claims to the £20,000 reward. The Board approved a £500 advance and Harrison began work on a successor to H1.

A long road
Over the next two decades, Harrison threw himself into perfecting his maritime clocks. By 1759, he had produced a contender for a full-blown sea trial to the West Indies. H3 had taken 19 years to make, weighed 60 pounds and had 753 separate parts. Sea trials were delayed by the Seven Years War against the French. When they eventually began, in 1761, Harrison had yet another timepiece to test. H4 was radically different to its predecessors. It was no bigger than a large pocket watch but, during the 81-day voyage to Jamaica, it lost just 5 seconds. Harrison ought to have been awarded his prize immediately but the Board of Longitude was heavily weighted towards astronomers and mathematicians who favoured the lunar method. They grudgingly awarded him £10,000 and demanded exhaustive tests to confirm the sea trials. In 1773, after King George III intervened, another payment of £8,750 was made. By this time, Harrison was an old man, exhausted by his efforts to satisfy the Board. He died in 1776.

Longitude for all
John Harrison's timepieces were wonderfully accurate but frighteningly expensive. In 1770, The Board of Longitude had ordered Larcum Kendall, a respected clockmaker to construct a copy of H4, to ensure that the original device hadn't been a fluke. Kendall's effort, known as K1, was a fine piece of work and it travelled with Captain Cook on his Pacific voyages, earning high praise from Cook for its accuracy. British clockmakers now had the bit between their teeth and, in the decades that followed, they produced several Harrison-style maritime clocks and watches. By 1800, British mariners had access to a ready supply of affordable timepieces. This technological advantage helped Britain's navigators conquer the oceans and consolidate its vast empire.