Thinking back before GPS…we know latitude can be readily determined using Polaris, the pole star, for locations above the equator. But, below the equator? There is no obvious marker on the celestial hemisphere.
Also, how could one measure longitude? At least, for latitude, it can be measured from the stars. But, I don’t see an obvious method for measuring longitude. If one could be in two places at the same time, one could measure a delta between the relative positions for the same star, using one position as a baseline. But, before two-way (remote) communication was possble, how was longitude found?
By clock. Determining local noon relative to Greenwich gives the info needed. The challenge of developing a reliable seagoing clock took many years and was the subject of intense competition (see marine chronometer for details) in the 1700s.
Prior to that, a lunar method could be used.
I believe that longitude was hard to measure accurately until the first clocks were designed (there was a big prize for the inventor) which could maintain accurate time over long sea voyages. With the clocks, one could compare when celestial objects touched the horizon with the predicted times in an almanac. Quite suddenly, maps began to look less “distorted” east-west.
As for the lack of a “south star” – interesting question! I think that if you observe a southern night sky for several hours, you can detect the approximate point around which the stars are “turning”, and there you have your south celestial pole. You will note where that point is relative to nearby constellations. (I think the Southern Cross points to it quite well, and you could count how many “cross lengths” there are from the constellation to the pole point). From then on, no matter where you go in the southern hemisphere, you would recognize that particular dark spot in the sky, and thus know your latitude.
(By the way, thanks to precession, this will be a creeping “problem” in the north, too, in a thousand years or so.)
The difficulty of accurately measuring longitude, as opposed to the ease of measuring latitude, is reflected in old maps of the time. A lot of them are pretty accurate along the north-south axis, while the east-west axis is bizarrely out of whack to modern eyes.
I had a little instruction in using a sextant to determine latitude, and we always used the sun at its highest point (LAN - Local Apparent Noon). On clear evenings, the woman who was teaching us would get out her sextant and take readings on some of the first stars to appear. I don’t even know what tables or math is involved, but you can find some information about your location from that.
I asked her about it; why those stars, and such. She said only a few stars come out while there’s still enough light to see the horizon. I’m not sure if Polaris is much help for determining location. By the time you can see it, it’s probably too dark to see the horizon through the sextant.
Draw a line through the southern cross length wise. Bisect the Pointers with another line. Where those two lines meet is south. I’m not sure if that method is accurate enough for navigation.
There are bubble sextants that measure angles relative to an internal “horizon” generated by a bubble. But these are typically much less accurate than using the real horizon (when it’s visible).
There’s a great animation here that shows a sextant and the image you would see when looking through it. (There’s a small flaw in that image; when the sun comes into view, you should only see the right half of it.) It does measure the angle above the horizon. (Strictly speaking, I think there are ways to use it horizontally, too, so it measures the angle between any two things you can see.)
I was amazed at how precise it was. In the calculations, we even had to account for how high our eyes were above the water (only about 20 feet.) I thought I was doing well to get within 20 miles. The second mate was within 1 (assuming you believe that fancy GPS thingy). That’s within a minute of arc.
I can’t think of any reason you couldn’t take a measurement with respect to level, instead of using the horizon, but you’d never be able to get that same precision on a moving ship. You could build something like that for use on land, but I think most land routes were well-known and mapped and weren’t the same sort of navigation problem.
All I know is from Sobel’s excellent book already mentioned, so go there for the straighter dope, but there are (and were) alternatives to the chronometer, involving nighttime astronomical sightings (‘moons of Jupiter’ was a favorite, I believe) and lots of complex calculations. These were used to some degree before accurate chronometers, but because of their complexity were more useful for fixed points on dry land than aboard a ship.
