NavList:

Paul, you asked:
"So when we learn plotting CN, you always start with a dead reckoning point. 
How was this done inland, with an artificial horizon in the days of Lewis & Clark? "

The artificial horizon really isn't the issue here. The methodology of land-based sights in the early 19th century was not much different from the "Age of Sail" methodology aboard ships at sea that we look at in some of my workshops.

You wait for local noon, and on land this is something that is more reliably predicted and planned for on the basis of compass direction. For example, on the L&C expedition up the Missouri, they could wait for the Sun to be exactly aligned at true South. Note though that this usually required a relatively simple check on the magnetic compass variation (also known as 'declination') to get the difference between true direction and compass bearing. With that in hand you shoot the Sun at noon, and then you have the latitude easily.

Later in the day, in the "afternoon", you shoot the Sun again. You can calculate the longitude from that, assuming you have a chronometer set to Greenwich Time (or time at any known location) or a non-chronometer watch set recently nearly to Greenwich Time using lunars. Note that the second Sun sight's longitude computation depends on the latitude, and if we have made good time in our canoes, then we might need to "DR" adjust the latitude. That's how a DR position gets into it --very similar to the situation for a navigator at sea.

Another trick: if we don't want to worry about adjusting the latitude for distance run, we can wait for the Sun to be very close to due west. Then the longitude determined has minimal dependence on the latitude. That works well in the Spring and Summer months (in Fall and Winter, the Sun sets before it reaches due west, so no help).

Frank Reed