NavList:

Bruce, you wrote:
"I have a question regarding resetting your clock when at a known location (or with a excellent fix , say from stars)."

Right away I would say there's a problem here. You say you have a known location or you have an excellent fix, maybe "from stars". These are very different things. When you're at a known location, you have some physical marker that identifies the place (hey look, it's the Empire State Building...) and you have a published list of latitudes and longitudes that includes that identifying point. When you have an excellent (celestial) fix, this can only be the case because you have an accurate chronometer or some other source of absolute time, in addition to the other elements of the toolkit of celestial navigation. It's very true that you can reset your clock from a known location, but you can't do it from an "excellent fix" since that's a circular line of reasoning.

You continued:
"Background: In August, a couple of hours before sunset when the sun was about due west, I pretended I did not know my time. I recorded an ordinary set of sights....really no special pains taken. Knowing my location, by initially guessing a time ( I knew it was a couple of hours before sunset) I quickly calculated my actual time. I recollect it was in error by a minute or less. If I did it again the next day, I suspect I could eventually get to 15 or 20 seconds error. Did mariners in the 1800s do this? So much easier than lunars and it can be done with any celestial body. This is nothing but a bit more than a “time sight” with some tabular interpolation."

It's certainly easier than lunars... but only if you already know where you are!! What you're describing is not similar to a time sight. It is a time sight. When you do a time sight, you convert an altitude of the Sun into local apparent time (inputs are latitude, declination, and observed altitude). You then convert that to local mean time by applying the equation of time. And then you compare that time with GMT. The difference is your longitude in time units:
  Longitude = GMT - LMT.

Finally you convert your longitude in time units to longitude in angular units using the usual rate of 15 degrees per hour (identical to four seconds per minute of longitude). Clearly if your GMT is uncertain in any way, you can take the equation for longitude and flip it around:
  GMT = Longitude + LMT.

Historically, it was quite common to check a chronometer at a known location using a time sight. In a remote location, this could be done by the navigator from the deck of his own ship. But of course it depended on the ability to identify some specific point and find it in a list of tabulated longitudes. In any well-travelled harbor, it was normal to assign this work to some shore observer. Then that "time authority" could fire a cannon, or raise a flag, or drop a time ball at some agreed-upon time. This was frequently at 13:00 local mean time. Then with the known time offset for that harbor, you would immediately have GMT to compare against the ship's chronometer. But whether the work's done aboard ship or by some "astronomer" on shore, it's fundamentally a time sight. Of course, an observer on shore is not limited to the use of a marine sextant. The timing on land can be much more accurate and may involve a transit observation instead of an altitude. But no matter how it's done, this is an astronomical observation that produces "sundial time", fundamentally identical to a time sight. This observation cannot replace a "lunar" any more than it can replace a chronometer. We're getting local time from the observation, not absolute time.

Frank Reed
Conanicut Island USA