NavList:

Bob Goethe, you wrote:
"in practice a lot of people continued sailing without chronometers (because of their expense) for a long time, and navigated via latitude-sailing."

Yes, and a great many of them also used lunars regularly, though regularly for lunars is not what many people imagine --once or twice a week is plenty. Lunars were much easier and much more common than many historical accounst imply. Their use at sea largely collapsed by 1850 even among frugal Yankee mariners, presumably (and this is still unverified by real primary source data, as far as I know) because second-hand chronometers became inexpensive enough for most navigators.

And you added:
"My understanding is that James Cook sailed to Tahiti using K1, the first copy of H4...and that the value of that chronometer exceeded the value of the rest of his ship."

Yes, definitely. That was one cheap ship! :) Seriously, it is quite true that the first copy, K1, was expensive to produce --Kendall was well-paid. Cook also had a number of chronometers by Arnold, of a cheaper, simplified design, but they were judged insufficient for long ocean voyages. Cook and his astronomers were able to make these judgments of the chronometers they carried only because they shot lunars extensively. Cook's vessel, Resolution (on the second voyage, when he was testing K2), really was a discount vessel. It was a collier (coal freighter), a rather average merchant vessel.

And you wrote:
"I said that Slocum's $5 clock was not much good for anything, navigationally, and that Slocum did a lunar - measuring the distance between the edge of the moon and a star - to work backwards to derive the time. But he talked about his long struggle with tables, and as nearly as I could tell he only did a lunar this once."

Spot on! And I'm glad to hear it. There's a nautical genie that keeps popping out of the bottle on this one. Many readers get the impression that Slocum was using lunars regularly. A dozen years ago, when I first suggested otherwise, there was much upset --a great disturbance in the Force. I managed to convince a few people that they were misinterpreting Slocum, but it was indirect reasoning. Shockingly, I eventually came across a smoking gun: Slocum himself wrote a letter home more than halfway through the voyage marveling at the accuracy of his navigation, despite having no chronometer, and, he said explicitly, having shot only one lunar in the entire voyage. Clearly, his principle methodology for his navigation was latitude by Noon Sun and longitude by dead reckoning. And that worked just fine.

And you wrote:
"He travelled across the Pacific by positioning himself on the right latitude and then staying there until he got to where he wanted to go."

Maybe not so much. Latitude sailing is a good idea, but if you can pick and choose when you sail, and if you're aware of other clues of the presence of land, it's not essential. 

You concluded:
"recreational sailors waited until Seiko introduced the quartz watch to REALLY be able to afford economical celestial navigation. This means that the golden age of celestial for small boat sailors lasted only from 1969 to sometime between 1992 and 2002, when GPS receivers became easily affordable: a golden age that lasted no more than 30 years or so."

Yes, that's a really interesting theory (and worth trying to test out with some hard data!). A confluence of factors made bluewater sailing more economical, more democratic in a fundamental way as the twentieth century rolled on, and as you say, a cheap chronometer-equivalent helped a lot. However, I would note that time-keeping at sea in the latter half of the twentieth century did not necessarily depend on a quality chronometer. It depended on good shortwave reception! From a certain point of view, the importance of the true chronometer was greatly reduced as soon as radio receivers could be carried on a majority of vessels. Could ocean-going yachtsmen afford radios by 1945? Maybe even by 1930?? A third-rate chronometer accompanied by a good shortwave is every bit as effective as a high-quality chronometer. Nonetheless, every bit helps, so I think it's possible that your theory has some validity. I wonder what data we might apply to it... :)

Frank Reed