In this video is explained that in November 1584 the width of the Atlantic ocean was discovered by comparing the hour when the eclipse of the moon started in England with the hour when the eclipse occurred in Virginia USA. But doesn't also the Sun arise on particularly times in England and USA to calculate the distance? Why was the eclipse of the moon used?

See min 22:00 https://www.youtube.com/watch?v=w6JZSzVcKgo

  • $\begingroup$ How do you propose to use "just the Sun" for determining the distance over the ocean?? $\endgroup$ Oct 21, 2017 at 17:38
  • $\begingroup$ Well if the Sun arise in England at 12:00 and in USA at 18:00 and you know the total circumference of the Earth than you can get pretty close. Or do I miss something more? $\endgroup$
    – Marijn
    Oct 21, 2017 at 18:05
  • $\begingroup$ Yes, that works if you have accurate clocks. But there were none in until the 18th century. I, in the USA, have to know the time in England when the sun rises here, not merely my own local time. The sun rises at the same local time everywhere at this latitude on this day...by definition of "local time". $\endgroup$ Oct 22, 2017 at 11:35
  • $\begingroup$ @Gerald Edgar: "accurate clocks" in 1584? What are you talking about? $\endgroup$ Oct 22, 2017 at 13:04

3 Answers 3


Perhaps the following quotewill be more, ahem, illuminating.

As any Eclipse chaser can attest, total solar eclipses are extremely rare from any given location. A spot on Earth may go centuries before totality graces the skies. In theory, a partial would be much more common. And of course, this would require something such as smoked glass to observe first contact. That leaves us with lunar eclipses. Any given eclipse can easily span more than one full hemisphere of the Earth. Also, while contact with the diffuse outer cone of Earth’s shadow, or penumbra, is rather subtle, contact with the inner core, or umbra, is not. Such an event can be defined by eye to a span of time of about five minutes. Up until the time of the invention of the telescope, eclipse timing was the only semi reliable way to judge longitude. The idea is simple; observe the local time of first contact with the umbra, then compare it with the known time from a separate fixed location. The Earth rotates 15 degrees per hour, (360/24=15) or 1 degree every 4 minutes. If umbral contact was said to begin at 2:05AM at a given location and you observed it to begin at 1:05AM, you are 15 degrees west of the first location. The Greeks first noted this during an eclipse on September 20th, 331BC. Alexander the Great in Mesopotamia recorded the eclipse as beginning two hours after sunset, while on the north shores of Africa, the city of Carthage it was seen to start at sunset. Ptolemy correctly deduced that Alexander’s position was 30 degrees or 2 hours east of Carthage. None other than Christopher Columbus was a student of the works of Ptolmey. He attempted to use this method to judge his position on his second and third voyage to the new world. The eclipses concerned were on September 14th, 1494 (part of saros cycle #119) and February 29, 1504 (saros cycle #109).

  • $\begingroup$ And yet, according to Wikipedia "Columbus therefore estimated the distance from the Canary Islands to Japan to be about 3,000 Italian miles (3,700 km, or 2,300 statute miles). The true figure is now known to be vastly larger: about 20,000 km". I wonder how this estimate could be so far off if these methods were so well known. Has this been Asked here? $\endgroup$ Feb 4, 2019 at 4:49

Read the interesting book Longitude: The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time to find why measurement of longitude simply by observing the sun (or the stars) was impossible. Until there were accurate clocks.


More precisely, they tried to determine the difference of longitude between two places. The difference of longitude is the difference in local times at the places. While local time is easy to determine by astronomical observations, there is no simple way of comparing time at different locations. Moon eclipse is one such mean. The principle was clear from the antiquity, and this was the principal method of practical determining of difference in longitude between distant places, until the discovery of Jupiter satellites, which have two advantages: their eclipses occur more frequently, and also their eclipse times are easier to predict with high accuracy.

EDIT. One cannot use a Sun eclipse for this purpose. The difference is that Moon's eclipse occurs at a definite time, which can be predicted. So if one observer observes it at his local time $T_1$, and another at his local time $T_2$ then the difference of longitude it $T_1-T_2$.

Sun eclipses occur at times which depend on locations.

  • $\begingroup$ This doesn't explain how they were able to compare the relative timings. Was it the position in the sky where the moon became fully eclipsed, compared with zenith angle of the moon? $\endgroup$ Oct 23, 2017 at 12:31
  • 1
    $\begingroup$ One does not need all these things. Moon eclipse occurs at a definite time, no matter from which place you look at it (provided that the Moon is visible from that place at the time of the eclipse). This time can be pre-computed. $\endgroup$ Oct 23, 2017 at 13:05
  • $\begingroup$ You don't even need to compute the time that precisely. Have a rough estimate so that you will be prepared. Then measure the position of the Moon at the moment of the eclipse. Compare notes later with the other party. $\endgroup$
    – timur
    Jan 3, 2020 at 2:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.