Talk:History of longitude

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Use of pendulum clocks[edit]

Huygens did make and use pendulum clocks. He also tried to use them to solve the longitude problem. Using a pendulum clock on land is the same as using them on the water - both are susceptible to motion-induced errors. He demonstrated that they could be built to be accurate but that the motion problem reduced their suitability.

I think the anonymous editor's changes are reasonable and more accurate that what was before. --Michael Daly (talk) 23:30, 10 November 2008 (UTC)hello people out in the world

Lunars or chronometers?[edit]

"It was less expensive to buy three chronometers, which could serve as checks on each other, than it was to acquire a high-quality sextant which was essential for lunar distance navigation."

This statement doesn't ring true. In order to fix his position, a navigator must in any case have an accurate sextant - the chronometer only gives him the current time so that he can calculate the co-ordinates of the object on which he is taking the sextant sight.

The sights for a position fix when GMT (or UT) is known require only a basic instrument. For this, through the middle of the 19th century, navigators used octants (usually made of wood) while the expensive, precision sextant (usually brass) was reserved for lunars. The sextant was, in fact, designed specifically for shooting lunars. Later in the 19th century, cheaper metal sextants eventually displaced the low-end octant but at the expense of accuracy. It is widely reported that sextants in the latter half of the 19th century were much lower in quality (see e.g. Lecky's "Wrinkles").

Secondly, I may well be wrong but I image a sextant, which is basically two (carefully calibrated and assembled) moving parts, two mirrors and a low-power (3x recommended) telescope, would have been cheaper to make than a chronometer.

See above. Also the parts of the sextant are not the critical components. It's the cutting and calibration of the scale as well as the centering of the index that makes a sextant accurate, and this was expensive. And it's worth remembering that this was market-driven. As lunars faded, the demand for very accurate sextants dropped. With declining demand, sextant quality. Without accurate sextants, lunars became less accurate. So lunars faded even faster... By the way, the "3x recommended" telescope is a modern perspective. In the 19th century, sextants for lunars were equipped even with 9x and 12x telescopes.

The three chronometer approach also has a fatal flaw - - if they show different times, how do you know which one is right?

You don't. So you have several options: 1) average them if the differences are small, 2) ignore one if it has suddenly deviated from the other two by a large change in rate, 3) assume that any of them might be right and navigate with an error band around your longitude that covers all three, 4) ask the first passing ship, "Hey, what is your GMT?" (this was quite common in the 19th century), 5) visit the nearest available port or known island at the earliest opportunity and compare the calculated longitudes against the charted longitude, 6) dust off the textbooks and shoot some lunars (as noted this was much less likely in the late 19th century, but it was always a last resort). And you'll note that even two chronometers would be enough to work with any of these solutions except the second. Also, three is not a magic number. Some ships carried five chronometers, or more. The point is that when you have multiple chronometers, the ensemble reveals its own errors rather than requiring an independent check by lunars (or other sources).

PS: If anyone has data on the price of chronometers and sextants at the end of the 19th century, I'd be interested. Scartboy (talk) 13:52, 5 September 2009 (UTC)

Many chronometer makers entered the business by that time, driving the prices down. They could make them fairly easily with mass production techniques (albeit low volume compared to other industries). Furthermore, navigators in merchant marine or fishing could make do with deck watches instead of chronometers. They could measure the rate of the watch (how fast or slow it was) and use the rate to correct the time. If the timepieces differed, the relative times could be compared on an ongoing basis to determine which one was showing an error (as opposed to its rate).
Sextants remained expensive and a high quality one quite expensive. While octants had dropped out of production by about 1880 or so (IIRC), they were adequate for use with a timepiece while inadequate for lunars. Hence, that sets the standard for what sort of sextant quality is required for lunars. --Michael Daly (talk) 03:14, 7 September 2009 (UTC)

Leibniz also claimed to have solved the longitude problem at the end of the 17th century....though I don't know the details, nor whether his insights had any merit worthy of including on this page. —Preceding unsigned comment added by (talk) 17:33, 24 June 2010 (UTC)

"Gradually but Rapidly"[edit]

The article's section "Lunars or chronometers?" includes the phrase, "gradually but rapidly." Could someone replace this self-contradiction with something unambiguous and meaningful?

During the mid- to late-19th century, affordable, reliable marine chronometers became available, gradually but rapidly replacing the method of lunar distance calculation. It became possible to buy three relatively inexpensive chronometers, serving as checks on each other, rather than acquiring a single (and expensive) sextant of sufficient quality for lunar distance navigation.

--Humanist Geek (talk) 19:36, 23 August 2010 (UTC)

Fixed. —Preceding unsigned comment added by (talk) 06:35, 11 October 2010 (UTC)

Longitude before the telescope[edit]

I`m in search for data on the accuracy of astronomic derived longitude (land based, I assume lunar eclipse or occulation) before the time of telescopes and marine dead reckoning accuracy (all times, all ships) -- Portolanero (talk) 14:35, 16 June 2011 (UTC)

One instance you may find interesting is the Islas de las Velas (Islands of the Sails) (now the Marianas Islands) mentioned in the Treaty of Zaragoza (1529) as being 17° east of the Mollucca Islands, but are actually about 17°21' east, which I thought was remarkably close by 16th-century standards. This longitude was determined by dead reckoning and solar observation by the navigators aboard Magellan's ships. See my discussion at Treaty of Tordesillas#Antimeridian: Moluccas and Treaty of Zaragoza. A source of raw data is Magellan's navigator's log book (1519) published by Lord Stanley of Alderley in his The first voyage round the world, by Magellan, London: Hakluyt, 1874, pp. 211–234. — Joe Kress (talk) 06:27, 17 June 2011 (UTC)
Thank you! I thought about this treaties too. The WP article is interesting. I see a big diference at the north of what the Cantino map shows as meridian and the modern map. That may be a result of AD 1500 dead reckoning. I found a source that modern DR systems on motorships had an error of 5 %. I expect much worse in 1500. I can not read your ref to Magellan's Voyage. This issue is not on the net. And this one seems to have another pagination. But anyway I`m not sure I could destill the DR error from this account. I`m not an expert here.
Just because of such treaties I expect they had some astronomical way in development or use. Lunar distance I estimate to +- 6°. Lunar eclipse to +- 1.5°. Lunar occulation could be better but its hard to estimate it. May depend on the stars magnitude. Thats why I look for real observations. There is evidence they did something but neither the method nor the results were much accepted. May be it was held secret. I assume they did at least lunar eclipse. But where is any direct record? Even a modern reenactment may help. To bad history of science has no much lobby. -- Portolanero (talk) 09:48, 18 June 2011 (UTC)
Most publications of Magellan's voyage only include Pigafetta's prose account. Virtually none except that by Lord Stanley of Alderley have the rather dry log book. You are probably a victim of Google Books' "selective geographic availibility" which limits views of a book to a single country or region such as North America or Europe, even for books in the public domain such as this one. Several other Google Books editions of the log book (one or more of which you may have access to) are: 1874, 1874, 2005, 2010.
The oldest coordinates are in Ptolemy's Geography, most of which were determined by pacing the distances. Muslim geographers improved these. During his fourth voyage, Columbus used a lunar eclipse predicted by Regiomontanus in his Ephemerides astronomicae to impress the natives of Jamaica [1]. — Joe Kress (talk) 03:28, 19 June 2011 (UTC)
Thanks a lot! Your suspicion on Google was right. I was not aware of this trick. I got advertisments to buy it from antiquarians. But the last two versions worked and I even found a full text here. It is very detailed and not easy to analyse. During the search I noted a paper "Magellan's Route in the Pacific" by George E. Nunn (1934) seems based on Alderley. I respect Nunn and he mentions maps but I can only see the first page. Maybe he already did it.
The coordinates of Ptolemy are somewhat odd. But he already mentions lunar eclipses. He only mentioned one, a very old and poor one. His book had such obvious poor data that it was almost a call for observations. -- Portolanero (talk) 10:51, 19 June 2011 (UTC)

Soory to be a pain but I'm having trouble reconciling just about everything in the "Ancient History" section with what is now known about the Antikythera mechanism. Bearing in mind the mechanism can hardly have sprung unbidden from the soil, those dates must be wrong. Furthermore, someone cleverer than I will have to explain how all the observations inherent in the mechanism's design were made without any form of telescope, surely.Drg40 (talk) 10:33, 14 May 2012 (UTC)

Longitude on Earth is totally different from the motion of Solar System bodies displayed by the Antikythera mechanism. Konwledge of that motion was developed by the Babylonians during the 1st millennium BC and transferred to the Greeks no later than 150 BC because that knowledge was used by Hipparchus. The only thing added by the Greeks was a geometrical model containing deferents and epicycles to the purely mathematical model used by the Babylonians. What is surprising about the device is that it mechanized this knowledge. It is an astronomical clock similar to those made during the Rennaisance (also before the telescope) but using a crank instead of a internal drive mechanism and one and a half millennia earlier. As such it simplifies complex astronomical motions now known because of the telescope to only those visible to the naked eye, although requiring hundreds of years of observation to quantify. The Babylonians began recording the positions of the Sun, Moon, and planets during the reign of Nabonassar about 750 BC. As such, the knowledge of Hipparchus as summarized by Ptolemy in his Almagest about AD 140 is highly complex. Conversely, the knowledge of the location of places on Earth contained in Ptolemy's Geography is quite rudimentary. — Joe Kress (talk) 07:42, 15 May 2012 (UTC)


"It became possible to buy two or more relatively inexpensive chronometers, serving as checks on each other" Never take two chronometers. One or three, but never two! — Preceding unsigned comment added by (talk) 01:12, 6 October 2011 (UTC)