34 COXCEPTIOXS OF COLUMBUS 



matters not. From Portugal he went to Spain. 

 Hence the point of departure for his voyage brought 

 him face to face with the same problem in na^*igating 

 the Atlantic westwards as he would haxe had start- 

 ing from Portugal. In the study of that Atlantic 

 there were three key points whence the problem as 

 Columbus faced it could be studied to better advan- 

 tage than elsewhere. These points were the Azores, 

 the Madeiras, and the Canaries. All three had been 

 known to the Portuguese for many years. The reader 

 is invited to study their position on the accompany- 

 ing map (PI. I).^ Their position with respect to the 



5 On PI. I the route of Columbus across the Atlantic and return has 

 been plotted according to the day's runs and courses as given in the ab- 

 stract by Las Casas of Columbus' log book (M. F. de Navarrete: Colec- 

 cion de los viages y descubrimientos que hicieron por mar los Espaiioles 

 desde fines del siglo XV, Vol. i, Madrid, 1825, pp. 1-166; Raccolta di 

 documenti e studi pubblicati dalla R. Commissione Colombiana pel 

 Quarto Centenario dalla Scoperta dell' America, Part L \'ol. i, Rome, 

 1892, pp. 1-119; English translation, with occasional errors in the figures, 

 in C. R. Markham: The Journal of Christopher Columbus During His 

 First Voyage, 1492-93, etc., Hakluyt Soc. Pubis., ist Series, Vol. 86. 

 London, 1893, pp. 15-193). The portions between Palos and the Canaries 

 and between the Azores and Palos have been omitted because the data 

 for these in the log book are insufficient. The day's runs on Sept. 26, 

 Oct. 9, and Oct. 11, which in the log book are given only as totals, have 

 been divided into the component parts estimated by G. V. Fox in the 

 table of distances and courses of the voyage (pp. 406-407 of The Log of 

 Columbus Across the Atlantic Ocean, 1492, Appendix D to his: An At- 

 tempt to Solve the Problem of the First Landing Place of Columbus in 

 the New World. Appendix No. 18 to £/. 5. Coast and Geodetic Survey Rept. 

 for 1880, Washington, 1882, pp. 346-411). Fox's allowance of 3 leagues 

 for departure from Gomera, Sept. 6-8, has also been used. Although a 

 certain leeway in the interpretation of the route is possible, the necessity 

 of fitting the outward and homeward tracks between known endpoints 

 makes it probable that any such reconstruction will in general be correct. 

 At all events the present reconstruction is sufficiently correct to show the 

 relation of the route to the physical conditions of the North Atlantic. 



