Page "Ptolemy" Paragraph 18

Because Ptolemy derived many of his key latitudes from crude longest day values, his latitudes are erroneous on average by roughly a degree ( 2 degrees for Byzantium, 4 degrees for Carthage ), though capable ancient astronomers knew their latitudes to more like a minute.

( Ptolemy's own latitude was in error by 14 '.

) He agreed ( Geographia 1. 4 ) that longitude was best determined by simultaneous observation of lunar eclipses, yet he was so out of touch with the scientists of his day that he knew of no such data more recent than 500 years before ( Arbela eclipse ).

When switching from 700 stadia per degree to 500, he ( or Marinos ) expanded longitude differences between cities accordingly ( a point first realized by P. Gosselin in 1790 ), resulting in serious over-stretching of the Earth's east-west scale in degrees, though not distance.

Achieving highly precise longitude remained a problem in geography until the invention of the marine chronometer at the end of the 18th century.

It must be added that his original topographic list cannot be reconstructed: the long tables with numbers were transmitted to posterity through copies containing many scribal errors, and people have always been adding or improving the topographic data: this is a testimony to the persistent popularity of this influential work in the history of cartography.