It is probable that there were Eskimo in the north-east of Newfoundland at that time, and that the natives may have been brought from thence or from southern Labrador.
Evidence of the Cantino map as to the Portuguese discoveries
Of all known maps the Cantino map undoubtedly gives the most complete and trustworthy representation of the Portuguese discoveries of 1500 and 1501 in the north-west; we know, too, that it was executed with an eye to these, at Lisbon, and immediately after the return thither of those who had taken part in the later voyage. We may consequently suppose that the cartographer availed himself of the sources then at his disposal. He may either himself have had access to log-books, with courses and distances, and to the original sketch-charts of the voyages, or he may have used charts that were drawn from these sources. But he used in addition maps and authorities of a more learned kind, as appears, for instance, in the legend attached to Greenland, where he speaks of the opinion of cosmographers, and says that this country is a point of Asia. It is clear, as pointed out by Björnbo [1910, p. 167], that Greenland was connected on the map with Scandinavia, which is called “Parte de assia,” but the upper edge of the map has been cut off, so that this land connection is lost,[351] as is the last part (asia) of the inscription on Greenland. The basis of this idea of a land connection must have been a map of Clavus’s later type; while the delineation of Greenland itself is evidently new. In fact, it is here placed for the first time very nearly at a correct distance from Europe, and with Iceland in a relatively correct position; and in addition to this it has been given a remarkably good form. If we assume that the variation of the compass was unknown, and that the coasts were laid down according to the courses sailed by compass as though they were true, then the southern point of Greenland comes just where it should, if the variation during the voyage from Lisbon averaged 11° west. The Portuguese flags on the coast indicate that the Portuguese sailed along the east coast of Greenland from north of the Arctic Circle of the map to past Cape Farewell (without landing, according to what the legend says), and its direction on the map is explained by a variation of about 14° west. The remarkably good representation of Greenland with the characteristic form of the west coast cannot possibly be derived from the Clavus maps, where Greenland is a narrow tongue of land with its east and west coasts running very nearly parallel. The west coast has been given a form approximately as though it were laid down from courses sailed with a variation increasing towards the north-west from 20° to nearly 30° (cf. [p. 371]). It is also characteristic that while the east coast is without islands, a belt of skerries is shown on the north along the west coast. It may seem a bold assumption to attribute this to pure chance and the caprice of the draughtsman, even though it may be pointed out that he has given the west coast of Norway a similar curved form with a belt of skerries outside (as on the Oliveriana map, [p. 375]). If the cartographer was acquainted with the representation of Greenland on the Clavus maps, the probability becomes still greater that he had definite authority for his west coast, since it differs from that of the Clavus maps. It is true that the Portuguese flags on the map and the statement in the legend that the Portuguese did not land on the coast do not seem to point to their having sailed any considerable distance to the north along the west coast, for otherwise there would doubtless be mention of this; but there may have been lost authorities for the Cantino map, which were based upon voyages unknown to us, as well as to the cartographer.[352]
If we may suppose that the lighter tone of the sea off the east coast of Greenland and over to Norway (on the original map) represents ice-floes, then this again gives evidence of a knowledge of these northern waters which we cannot assume to have been derived merely from Portuguese voyages on which the east coast of Greenland was sighted; it must have had other sources, unknown to us.
Construction of the Cantino map
There can be no doubt that the “Terra del Rey de portuguall” of the Cantino map is the east coast of Newfoundland, which, through the variation of the compass being disregarded, is given a northerly direction. If we draw the east coast of Newfoundland from Cape Race to Cape Bauld on approximately the same scale as that of the Cantino map, and turn the meridian to the west as far as the variation may have been at that time (about 20° at Cape Race, and 4° or 5° more at Belle Isle Strait), we shall have a map (see [p. 364]) the coast-line of which bears so great a resemblance to that of the Cantino map that it is almost too good to believe it not to be in part accidental (the Newfoundland coast on Reinel’s map is also very nearly the same as that of the Cantino map). The resemblance is so thorough that we might even think it possible to recognise the various bays and headlands; but perhaps a part of the southern coast of Labrador has been included in the Cantino map. According to the scale attached to the map, in which each division represents fifty miglia, the distance between the south-eastern point of the country and the northern Portuguese flag is seven hundred miglia, which thus corresponds to the six hundred or seven hundred miglia that Pasqualigo says the Portuguese sailed along the coast. If we divide the map into degrees according to the distance between the tropic and the Arctic Circle, the extent of the country will be about eleven degrees of latitude. On Reinel’s map the length of Newfoundland from north to south is between ten and eleven degrees of latitude. The distance from Cape Race to Belle Isle Strait corresponds in reality to about 5½°, that is, fairly near the half.
Reconstruction of an equidistant chart on which the coasts are
laid down from magnetic courses without regard to the variation
Both Greenland and Newfoundland lie too far north on the Cantino map. The southern point of Greenland lies in about 62° 20′ N. lat., instead of 59° 46′, while Cape Race, the south-eastern point of Newfoundland, lies in about 50° N. lat., instead of 46° 40′. It is unnecessary to assume that the too northerly latitude of Greenland is derived from the Clavus map, where its southern point lies in 62° 40′ N. lat., since a natural explanation of the position both of this point and of Cape Race is provided by the way in which the Cantino map is drawn. It is, in fact, an equidistant compass-chart, which takes no account of the surface of the earth being spherical and not a plane, and on which the courses sailed have been laid down according to the points of the compass, presumably in ignorance of the variation of the needle. If we try to draw a map of the same coasts in the same fashion, using the correct distances, and taking the courses as starting from Lisbon, and the variation to be distributed approximately as given on [p. 308],[353] we shall then get a map in its main outlines as here represented. The southern point of Greenland comes in about 62° 20′, or the same as on the Cantino map, and Cape Race comes still farther to the north than on it. The distance from Lisbon to Greenland is almost exactly the same on both maps, and this seems to point to remarkable capabilities of sailing by log and compass, while, on the other hand, astronomical observations were probably not used. The distance between Lisbon and Newfoundland (Terra del Rey de portuguall) is on the Cantino map a little longer than reality,[354] and the southern end of the latter is brought so far to the south that it would correspond to an average variation of about 4° west, instead of 10°, during the voyage from Lisbon. Newfoundland accordingly comes farther west in relation to Greenland, and its southern end farther south than it should do on a map constructed like this one. But we do not know whether the course from which the position of Newfoundland is laid down was taken as going directly to that country from Lisbon; perhaps, for instance, it went first up into the ice off Greenland, and in that case a greater error is natural. If we lay down the West Indian islands (and Florida) on our sketch-map according to the same method, we shall get them in a similar position to that of the Cantino map, except that there they have a far too northerly latitude, and the distance from Lisbon is much too great; but this is due to the Spanish maps which served as authorities; for we know that even Columbus was guilty of gross errors in his determination of latitude,[355] and on La Cosa’s map they lie for the most part to the north of the tropic.
Variation in the Portuguese representation of Greenland