By: Henrique Leitão and Joaquim Gaspar
From: Universidade de Lisboa
At: Instituto de Investigação Interdisciplinar, Anfiteatro
The celebrated cartographical projection first used by Gerard Mercator (1512-1594) in his map of 1569 is justly considered a landmark in the histories of cartography and of marine navigation. Mercator left no description of the method he used for his map and speculation about the possible construction methods has been lively for more than a century. However, no conclusive agreement has been achieved among historians so far.
Recently, we proposed a novel â€“ and in our view, final â€“ solution to this question. We performed a detailed analysis of the errors affecting the graticule of Mercator's map. From the error distribution we identified on the map we concluded not that none of the methods hitherto suggested can be accepted as an explanation of Mercator's procedure. Furthermore, we identified a numerical procedure (using a table of rhumbs) that reproduces almost exactly the measured errors. We also showed that the construction of such a graticule with reference to a table of rhumbs is a trivial process. Finally, to complement our numerical study, we traced the history of tables of rhumbs â€“ a concept originally due to the Portuguese mathematician Pedro Nunes (1502-1578) â€“ and showed that, for historical reasons, using such a numerical technique was the most likely method for Mercator to use.