One of history's coincidences allows me to use this title with impunity. One slight admission, of course, is that I'm talking about a different John Harrison. It was, however, a John Harrison who was working in the 1760s and who really was an early practitioner of
lunar distances at sea.
This John Harrison, about whom disappointingly little is known, sailed as purser on HMS
Dolphin's circumnavigation of 1766-68 under the command of
Samuel Wallis. What we do know is that he had an interest in astronomy and mathematics and made accurate determinations of the positions of islands discovered during the voyage. These included
Tahiti, which Wallis named King George's Island after it was first sighted in June 1767 (the
Dolphin being the first European vessel to do so). Wallis noted in his journal that Harrison had established the island's position by, ‘taking the Distance of the Sun from the Moon and working it according to Dr Masculine’s Method which we did not understand’ (Harrison presumably had a copy of
Maskelyne's 1763 work on lunar distances,
The British Mariner's Guide). This was a fortunate discovery, as the island was in the region that Maskelyne had prescribed as favourable for observations of the next transit of Venus, due in 1769. Indeed, this was why
Cook was sent to Tahiti on his
first circumnavigation of 1768-71. On that voyage, Cook judged that Harrison's position had been correct to within half a degree, which sounds pretty good.
Spending a morning at Cambridge University Library a few weeks ago, I came across an interesting related document. This was a letter from the same John Harrison to
William Wales, dated May 1787 (
RGO 14/187 item 13). Evidently Wales had been hoping to get more information about some eclipse observations made during the
Dolphin's voyage and was hoping Harrison could add something. Sadly, all Harrison could say was that, 'I am really ashamed to say I have not a paper by me relative the Voyage you mention, having lost every mathematical Book & paper I was master of'. Harrison does, however, confirm that all the observational results given in the surviving logs were 'per Quadrant', meaning that they had not been corrected for parallax and other errors. Harrison and his fellow mariners were clearly happy to let others do the tedious mathematics of reducing the observations, which was also one of the downsides of the lunar distance method, as Harrison presumably knew well from his longitude determinations.