l8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 59 



and the name also have been explained by the Atlantis legend, since 

 Plato's description corresponds with it more or less. But there is no 

 obvious reason why this influence should have been less potent in the 

 fourteenth century, .the quite numerous maps of which have no draw- 

 ings of Antillia. 1 Humboldt argued against both suppositions and 

 thought the name derivable from the Arabic " Al Tin," the serpent 

 or dragon, a reminiscence of the terrors of the Sea of Darkness. In 

 support of his contention he refers to the Island of the Dragon and 

 like items. It is certainly true that Edrisi has a passage concerning 

 that destroyer, killed, as he says, on one of the Azores by Alexander 

 the Great ; that the Pizigani's kidnapping monster is distinctly labeled 

 " a dragon " and that even the much later Olaus Magnus 2 decorates 

 one issue of his history with a pictured saurian having a serpent's 

 tail, in the act of dragging a sailor from a ship's deck to its lair on 

 some rocky Atlantic shore. Evidently huge reptiles of the lizard 

 kind were associated in human minds for five or six centuries with 

 the perils of westward navigation. This of course may mean no more 

 than a play of fancy about memories of crocodile-haunted African 

 rivers ; though it may also conceivably record impressions left by far 

 western islands where similar forms were at least equally common. 

 D'Avezac, 3 reviewing the matter of etymology in 1845, dissented from 

 Humboldt's hypothesis ; which does not seem to have been taken up 

 zealously by any advocate, notwithstanding the very great eminence 

 of its author. Perhaps it has been regarded as ingenious, rather than 

 perfectly reliable, for the transformation of Altin into Antillia is not 

 adequately explained. 



A more plausible conjecture, probably the most nearly convincing 

 one thus far offered, makes up the name in Portuguese from Ante or 

 Anti (before or opposite) and ilha Island. On some maps the latter 

 word regularly becomes ilia for example that attributed to Zuan 

 da Napoli, 4 already mentioned. By either spelling, the pronunciation 

 in full would presumably be Anteillia or Antiilia, readily compressed 

 to Antillia, after the manner of all languages when two similar vowels 

 come together. Obviously this derivation has the advantage of sim- 

 plicity and the case as to meaning is equally good. Divers early maps 

 as Battista Beccaria (Becharius) 1435, Bianco 1436, Pareto G 1455, 

 Roselli 1468, Bertran 1489, and Benincasa, 1482 show Antillia, 



1 Jomarcl: Atlas, Plate 11', Pizigani Map of 1367. An obscure Latin inscrip- 

 tion on it contains, however, the word Atullae or Atillie, identified with 

 Antillia by Kretschmer and others. 



2 J. Winsor: Narr. and Crit. Hist, of America, vol. i, p. 74; Tillinghast's 

 Monograph. 



3 Les lies Fantastiques, p. 27. 



* Kohl's collection of maps in Library of Congress. 



5 K. Kretschmer: Die Entdeckung Amerikas, Atlas. Tafel 4. 



