COLUMBUS' DISCOVERY OF VARIATION. 201 



ing of the Blanco chart, which modern research has since 

 proved to have been misinterpreted. The fact that a sim- 

 ilar dictum to that of Humboldt is advanced by Washing- 

 ton Irving in his fascinating life of the Admiral has done 

 much to place the matter apparently beyond dispute ; but 

 an impartial study of the history of the rise and progress 

 of magnetic knowledge up to the time of Columbus, and 

 of the condition of it during his life, and a recognition of 

 the fact that much important data underlying such history 

 has been made known since both Humboldt and Irving 

 wrote, indicate the need for a revision of their verdict. 



Little weight can be given to the argument that the first 

 freely suspended magnetic needle certainly showed varia- 

 tion, as did all later ones when influenced by the earth's 

 field, and that therefore the phenomenon was always open 

 to observation. Unfortunately many a physical effect has 

 thus presented itself for ages to the perception of man- 

 kind nay, forced itself under the very eyes of the keen- 

 est investigators without gaining recognition, or adding 

 in the slightest to the world's stock of knowledge, until 

 suddenly hailed as a great discovery. Moreover there 

 were cogent reasons why, even if navigators had noted an 

 aberration of the needle, they would have been likely to 

 ascribe it to other causes than the true one, and so have 

 failed to recognize the real variation at all. . 



Thus, in May, 1496, when the Genoese and the Flem- 

 ish compasses on the ships of Columbus were found to 

 disagree, one varying to the northwest and the other in- 

 dicating the star, Columbus himself concludes the reason 

 to be the difference in the magnets with which the needles 

 were rubbed. In such rudely constructed instruments as 

 then existed, it was equally possible to have assigned the 

 errors to difference in shape of the needles, or weakness 

 of magnetization, while it is not at all unlikely that both 

 their form and treatment resulted in the production of con- 

 sequent poles, which imported into them still further error. 

 There was much better reason, therefore, for the European 



