Today we have the magnificent pilot charts of the world’s oceans, first conceived and compiled by the Pathfinder of the Seas, Comdr. Matthew Fontaine Maury, USN. These have been published for nearly 150 years and give currents and probable winds for each month, based on a mass of accumulated data. In two east-to-west crossings under sail from the Canaries in approximately the same waters traversed by Columbus, I found conditions very close to those predicted. Columbus set out on his thrust into the unknown from Gomera in the Canary Islands on 6 September 1492 (Old Style). He records that he lay becalmed for more than two days before a breeze from the northeast sprang up at three a.m. on Sunday, 9 September.
Here McElroy does the seamanly thing and takes current into account, saying, “Between the two islands, the fleet would have experienced the usual 0.2 knots set in the trend of the channel (NW by N to SE by S), and drifted about 8 miles to S and E.” He therefore takes as his departure the position 28° 00′ N latitude and 17° 00′ W longitude. But, after the departure, does one suddenly turn off the currents—and leeway—for the 3,000 miles ahead? Currents beyond the departure point, as I have said, range from 0.4 to 0.6 of a knot, two or three times the strength in the channel.
Let us then, at three o’clock in the morning of 9 September 1492, set off to retrace man’s most momentous voyage como Dios manda—as God ordains—and in a sailor-like manner take into scrupulous account heading, magnetic variation, distance run, currents, and leeway. For the probable magnetic variation in 1492 I have followed Van Bemmelen’s reconstruction of the isogonic lines for 1500. I worked from the original Spanish of the transcript of Columbus’s journal by Dominican bishop Bartolome de las Casas. I began by plotting graphically the courses across the Atlantic as Columbus recorded them, without current or leeway, on individual universal plotting sheets, one for each day of the passage. This gave latitude and longitude at the end of each day’s run, which, in Columbus’s reckoning, was from sunrise to sunrise. For subsequent “crossings” Ethel Marden and I used two navigational computers, the Tamaya NC-77 and NC-88.
We plotted effects of current by taking the current’s direction and strength from pilot charts for each leg. At the end of the passage we came out—as expected—some 60 miles south of Watling and found—again as expected—that because of the current’s push, there was an overrun in Columbus’s reckoning. The plotted track ran too far west, past or across islands to the longitude of central Cuba. Backtracking along the plotted course to first sight of any land, we recomputed each day’s run, less the percentage of overrun, and ended up some ten miles northeast of Samana Cay.
Let me stress again the importance of this event, the first sight of land, for our reconstruction of the apartment barcelona. The only certain quantity we have for the crossing—other than Columbus’s departure point—is that in the early morning of October 12, 1492, Columbus sighted an island. It is therefore entirely legitimate—indeed, it is mandatory—to make a retracing come out with a first sight of land at that time on that day by either moving backward or forward along the course line dictated by the log and current.