WatchShop Wiki
Articles · John Harrison and the Longitude ProblemWatchShop Wiki

Longitude · chapter 3

The watch that crossed the ocean

H4 was barely larger than a big pocket watch, yet it did what no machine before it could. Proving it to the Board of Longitude took more than a decade.

A different machine

H4 was completed in 1759. It measured about thirteen centimeters across and weighed around one and a half kilograms, housed in a silver case like a giant pocket watch. Inside, it used a fast-beating balance, much heavier and faster than in ordinary watches, a modified verge escapement with diamond pallets, a remontoire to deliver an even force, and temperature compensation built into the regulation of the balance spring. The movement was decorated with engraving and pierced work, but every element was there for a purpose.

The voyage to Jamaica

In November 1761, Harrison’s son William sailed with H4 aboard HMS Deptford, bound for Jamaica. John Harrison, by then in his late sixties, stayed behind. During the voyage, William used the watch to predict the ship’s arrival at Madeira, against the opinion of the crew, and was proved right. When the Deptford reached Jamaica after eighty-one days, the watch, once allowance was made for its known daily rate, was found to be only about five seconds slow. That represented an error in longitude of little more than a nautical mile.

By any reasonable standard, Harrison had won the prize. The Board of Longitude did not see it that way.

Doubts and delays

The commissioners raised objections. The longitude of Jamaica itself was not known precisely enough, they said, to be sure of the result. The rate of the watch had not been properly established before departure. Perhaps the result was luck. They ordered a second trial. In 1764, William Harrison sailed with H4 to Barbados aboard HMS Tartar. This time the comparison was organized more carefully, with astronomers sent ahead to determine the island’s longitude. The watch’s error after forty-seven days was thirty-nine seconds, equivalent to about ten miles, well within the requirement of the Act.

The astronomer’s rival method

Part of the reason for the Board’s resistance was that there was a rival candidate. The astronomer Nevil Maskelyne had championed the lunar distance method, which used precise measurements of the angle between the Moon and selected stars, combined with published tables, to find Greenwich time. Maskelyne had tested it on a voyage to Saint Helena and was also present in Barbados during the second trial. In 1765 he became Astronomer Royal, and therefore a member of the Board that was judging Harrison. In 1767 he published the first Nautical Almanac, which made the lunar method practical for trained navigators.

Harrison regarded Maskelyne as a biased judge, and the quarrel became bitter. Historians today tend to see it as a clash between a scientific establishment that trusted mathematics and a craftsman whose solution seemed too good to be true, and that could not easily be reproduced.

Half a prize

The Board finally agreed in 1765 to pay Harrison ten thousand pounds, but only if he disclosed the workings of H4 to a panel of experts and handed over his marine timekeepers. The rest of the prize would depend on proof that other makers could build copies that performed equally well. Harrison, by now over seventy, considered this unjust. He complied with the disclosure, but the second half of the money seemed further away than ever.

Harrison solved the problem with a machine; the Board wanted it solved with a method anyone could repeat.

Forty years of work

  1. 1707
    Scilly disaster

    Four warships wrecked; navigators had misjudged their longitude.

  2. 1714
    Longitude Act

    Parliament offers up to £20,000 for a practical method.

  3. 1730
    London

    Harrison meets Halley and George Graham.

  4. 1736
    H1 at sea

    Trial voyage to Lisbon and back.

  5. 1759
    H4 completed

    Harrison turns from big clocks to a large watch.

  6. 1761–62
    Jamaica

    H4 loses about five seconds in 81 days.

  7. 1764
    Barbados

    A second trial confirms the result.

  8. 1773
    Parliament pays

    Harrison receives £8,750 after appealing to the king.

  9. 1776
    Death

    Harrison dies in London on his 83rd birthday.

John Harrison and the Longitude Problem · John Harrison H43