For the northern coast of South America La Cosa and Cantino’s draughtsmen seem to have had different authorities. La Cosa attaches forty-five names to that coast: Cantino only twenty-nine; and only three of them are common to the two.[417] Harrisse argues from the failure of the La Cosa map to give certain intelligence of the Atlantic coast of the United States (here represented in the north and south trend of shore, north of Cuba), that there was existing in October, 1500, at least in Spanish circles, no knowledge of it,[418] but that explorations must have taken place before the summer of 1502 which afforded the knowledge embodied in this Cantino map. This coast was not visited, so far as is positively known, by any Spanish expedition previous to 1502. Besides the eight Spanish voyages of this period (not counting the problematical one of Vespucius) of which we have documentary proof, there were doubtless others of which we have intimations; but we know nothing of their discoveries, except so far as those before 1500 may be embodied in La Cosa’s chart.[419] The researches of Harrisse have failed to discover in Portugal any positive trace of voyages made from that kingdom in 1501, or thereabout, records of which have been left in the Cantino map. Humboldt had intimated that in Lisbon at that time there was a knowledge of the connection of the Antilles with the northern discoveries of Cortereal by an intervening coast; but Harrisse doubts if Humboldt’s authority—which seems to have been a letter of Pasqualigo sent to Venice, dated Oct. 18, 1501, found in the _Diarii_ of Marino Sanuto, a manuscript preserved in Vienna—means anything more than a conjectural belief in such connection. Harrisse’s conclusion is that between the close of 1500 and the summer of 1502, some navigators, of whose names and nation we are ignorant, but who were probably Spanish, explored the coast of the present United States from Pensacola to the Hudson. This Atlantic coast of Cantino terminates at about 59° north latitude, running nearly north and south from the Cape of Florida to that elevation. Away to the east in mid-ocean, and placed so far easterly as doubtless to appear on the Portuguese side of the Line of Demarcation, and covering from about fifty to fifty-nine degrees of latitude, is a large island which stands for the discoveries of Cortereal, “Terra del Rey du Portuguall;” and northeast of this is the point of Greenland apparently, with Iceland very nearly in its proper place.[420] This Cantino map, now positively fixed in 1502, establishes the earliest instance of a kind of delineation of North America which prevailed for some time. Students of this early cartography have long supposed this geographical idea to date from about this time, and have traced back the origin of what is known as “The Admiral’s Map”[421] to data accumulated in the earliest years of the sixteenth century. Indeed Lelewel,[422] thirty years ago,