chapter xvii., about the possibility of sailing round the world, which

he declares to be practicable, and if ships were sent out to explore the world, one could sail round the world, both above and below. He says that when he was young he heard of a man who set out from England to explore the world, and who went past India and the islands beyond it where there are more than five thousand islands, and so far did he travel over sea and land that he finally came to an island where he heard them calling to the ox at the plough in his own language, as they did in his own country. This island afterwards proved to be in Norway.[262]

Clavus's assertion that he himself saw ("ut uidi") Karelians in Greenland is impossible. As it is expressly stated that there was land at the North Pole, and as it is not mentioned that these Karelians had hide-boats like the Pygmies, the meaning must be that their armies came marching by the land route, which, of course, is an impossibility, which, if he had been in Greenland, would make him a worse romancer than if we suppose his "ut uidi" to mean that he had seen something of the sort stated in a narrative; but even this may be doubtful. In the Bruges itinerary [cf. Storm, 1891, p. 20] or some similar older authority, which we know he may have used, he may have seen "Kareli" beyond Greenland spoken of as "in truth a populus monstrosus." We have already said that on the maps accompanying Marino Sanudo's work he may have seen "Kareli infideles" marked on the mainland to the north-east of Norway, or even on an island out in the northern sea, and he would then naturally have connected the Karelians of the itinerary with these Karelians north of Norway. If we add to this that on the Medicean map of the world he saw the mass of the continent extending from Scandinavia and the peninsula corresponding to Greenland, northwards into the unknown, and that in the "Rymbegla" tract he saw mention of land at the North Pole--then, indeed, his whole statement seems to admit of a perfectly natural explanation.

His lack of knowledge of the conditions in Greenland appears again in his speaking of Pygmies and Karelians as two different peoples, one apparently on the sea, and the other marching in armies on land; and in his mentioning hide-boats as something peculiar to the former in the fabulous northern country, while he does not say that the Karelians in Greenland had boats or went to sea. If he had only spoken to people who had been in Greenland, he could hardly have avoided hearing of the Skrælings who come to meet every traveller in their hide-boats.

[Sidenote: Clavus's west coast of Greenland taken directly from the Medici map]

It is an important difference between Clavus's first and second maps (and also between his first and second texts) that on the latter Greenland is given a west coast. Its form bears an altogether striking resemblance to the west coast of the corresponding peninsula on the Medicean mappamundi, so that there can be no doubt that this coast is copied from it.[263] This is notably the case if we confine ourselves to Björnbo and Petersen's reconstruction of the coast after the text of Clavus, from which it appears plainly enough that there are the same number of bays as on the Medici map; they are closest together near the southern point of the country; then come two larger bays to the north, then a very broad bay, longer than the two others together, and then a straighter coast-line to the north of that (cf. p. 236). The east coast of Greenland has in part been provided with corresponding bays, although this coast is almost straight on the Medici map; but this answers to the north coast of Scandinavia on the Nancy map having very nearly the same indentations as the south coast. In taking the Medici map as the foundation of Clavus's Greenland coast we also have a natural explanation of the relation between his distribution of names on the east coast and the west. In his later text it is striking that his description of the east coast of Greenland does not reach farther than to his "Thær promontorium" in 65° 35', while the description of the west coast goes as far north as 72°. This might seem to be connected with real local knowledge, since the latitude 65° 35' on the east coast agrees in a remarkable way with the latitude of Cape Dan, 65° 32', where the coast turns in a more northerly direction. To the north of this the coast is usually blocked with ice, and this place has therefore frequently been given as the northern limit of the known east coast, and probably it was there that the Icelanders first arrived off the land on their voyage westward to the Greenland settlements. But this is one of those accidental coincidences that sometimes occur, and that warn us to be careful not to draw too many conclusions from evidence of this nature.[264] We find the explanation in the Medici map (p. 236), where the east coast of the peninsula corresponding to Greenland does not go farther north than to about the same latitude as the promontory on the south side of the broad bay already referred to on the west coast, which promontory Clavus calls "Hynth" ["Hyrch"]; it lies in 65° 40'. As Clavus's coast from this point of the east coast northward had no map to depend on, he did not venture to go farther in his description this time, though in the Nancy text he goes to 71° with his northernmost cape.

The Medicean map of the world gives us at the same time a simple explanation of Clavus's designations for the two most northerly points on the west coast of Greenland. If we confine ourselves to the scale of latitude for the Medici map, which, as stated above (p. 259), we have found by using Ptolemy's latitudes for more southern places on the map (Gibraltar and Brittany), and which is inserted in the left-hand margin of the reproduction, p. 236, we shall find the following: just at the spot of which Clavus declares: "New, the uttermost limit of the land which we know on this side, lies in 70° 10',"[265] the heavy colouring of the land on the Medici map comes to an end (judging from the photograph in Ongania, Pl. V.). Farther to the north extends the coast of the lightly coloured mass of land; but just at this point, in 72°, where Clavus has his "ultimus locus uisibilis" [last point visible][266] this coast-line disappears into the oblique frame which cuts off the upper left-hand corner of the map. The agreement is here so exact and so complete that it would be difficult to find any way out of it.

[Sidenote: The position of Iceland]

Björnbo and Petersen have asserted that Iceland, on the later map and in the Vienna text, has been given a position more in agreement with the sailing directions than on the Nancy map. I cannot see the necessity for this supposition, as it has almost exactly the same position in relation to the southern point of Greenland and to Norway in both works; the chief difference is merely that the longitude of all three countries is made 3° farther east in the later work (and the latitude of the southern points of Iceland and Greenland is put somewhat farther south), and that the east coast of Greenland has a more oblique north-eastward direction than the corresponding north-east coast on the Medici map, with the direction of which the Nancy map agrees fairly well. In this way it is brought nearer to Iceland; but that this should be due to a knowledge of the sailing directions seems very uncertain, and is not disclosed, so far as I can see, elsewhere in the later work. The only things I have found which might possibly point to northern authorities having been consulted since the production of the Nancy work, are the accurate latitude of Trondhjem, already referred to, and the island of "Byörnö" between Iceland and Greenland. The latter might be the Gunnbjörnskerries (or Gunnbjarnar-eyar) mentioned, amongst other places, in Ivar Bárdsson's description of Greenland; but the abbreviation of the name is curious. Perhaps the island may be due to some oral communication, or an erroneous recollection of something the author may have heard of in Denmark in his youth.

[Sidenote: Clavus's merits]

On the whole we shall be compelled after all to detract considerably from Claudius Clavus's reputation as a Northern traveller and cartographer. His journey did not extend farther north than the Danish islands, and perhaps Skåne. On the other hand, he was in Italy, where he drew his maps or had them drawn, and where he also found his most important authorities. His chief merit as a cartographer is that he is the first we know of to have adopted Ptolemy's methods, and that he gave the name of Greenland to the westernmost tongue of land in Norway on the Medicean mappamundi, and altered this a good deal with the help of other compass-charts and Vesconte's mappamundi, to make it agree better with the ideas of the North which he may have acquired to some extent in his youth through legendary tales, and later through Saxo and other writers.

[Sidenote: Clavus's influence on later cartography]

[Sidenote: Nicolaus Germanus, circa 1460-1470]

Claudius Clavus's later map of the North exercised for a long period a decisive influence on the representation of Scandinavia and to some extent of Greenland. This was chiefly due to the two well-known cartographers, Nicolaus Germanus and Henricus Martellus.[267] The former must have become acquainted with Clavus's map soon after 1460, and included copies of it in the splendid MSS. of Ptolemy's Geography which proceeded from his workshop at Florence. In these copies, of which several are known (cf. p. 251), he has redrawn Clavus's map in the trapezoidal projection invented by himself, whereby his Greenland has been given a more oblique position than the Greenland of the original map and the corresponding peninsula on the Medici map. He also introduced this Greenland into his map of the world [cf. J. Fischer, 1902, Pl. I., III.; Björnbo, 1910, p. 136]; but, in order to make it agree better with the learned mediæval view of the earth's disc surrounded by ocean, he surrounded it by sea on the north, so that it came to form a long and narrow tongue of land projecting from northern Russia, instead of the northern mass of land extending to the North Pole according to Clavus. But this long peninsula does not seem to have entirely satisfied this priest's erudite ideas of the continent, and on later maps (which were printed after his death in the Ulm editions of Ptolemy of 1482 and 1486) he shortened it so much that it became a rounded peninsula to the north of Norway, with the name "Engronelant,"[268] and at the same time he moved Iceland out into the ocean to the north-west. This apparently quite arbitrary alteration may perhaps be due to a desire to bring the map as far as possible into agreement with the learned dogma of the continent [cf. Björnbo, 1910, pp. 141, ff.]; but older conceptions of Greenland may also have contributed towards it [cf. J. Fischer, 1902, pp. 87, ff.]. We have already seen that Adam of Bremen regarded Greenland as an island "farther out in the ocean opposite the mountains of Suedia" (see vol i. p. 194), and in his additions to the copy of Ptolemy, Cardinal Filastre (before 1427) states that Greenland lay to the north of Norway; we find the same view in the letter of 1448 from Pope Nicholas V. (see above, p. 113).[269] It is also somewhat remarkable that on the Genoese mappamundi of 1447 (or 1457) there occurs a peninsula north of Scandinavia just at the place where Clavus's Greenland should begin (see p. 287).[270] On Fra Mauro's mappamundi (1457-59) there are several peninsulas to the north of Scandinavia, some of which proceed from Russia (see p. 285).

[Sidenote: Henricus Martellus, circa 1490]

The cartographer Henricus Martellus, who succeeded Nicolaus Germanus, again adopted Clavus's form of Greenland, wholly or in part, on his maps dating from about 1490.

In this way there arose on the maps of the close of the Middle Ages two types of the North: one with Greenland in a comparatively correct position to the west of Iceland, though far too near Europe and connected therewith, and another type with "Engronelant" as a peninsula to the north of Norway. The latter remained for a long time the usual one in all editions of Ptolemy, in other cartographical works and on many globes. After the rediscovery of Greenland we even get sometimes two delineations of this country on the same map, one to the north of Norway and the other in its right place in the west.

[Sidenote: Illa verde]

Greenland seems to have been given a wholly different form on a Catalan compass-chart from Majorca, of the close of the fifteenth century, where in the Atlantic to the west of Ireland and south-west of Iceland ["Fixlanda"] there is an island called "Illa verde" [the green isle]. It seems, as assumed by Storm [1893, p. 81], that the name must be a translation of Greenland, which is called in the Historia Norwegiæ "Viridis terra." The representation of Iceland ["Fixlanda"] on this map is incomparably better than on all earlier maps, and gives proof of new information having come from thence. As the place-names point to an English source, it is possible that the cartographer may have received information from Bristol, which city was engaged in the Iceland trade and fisheries, and his island, "Illa verde," may be due to an echo of reports about the forgotten Greenland in the west. It is worth remarking that the island is connected with the Irish mythical "Illa de brazil," which lay to the west of Ireland and which appears in this map twice over in its typical round form (cf. above, p. 228).[271] If we remember that this happy isle is in reality the Insulæ Fortunatæ, and that in the Historia Norwegiæ (see above, p. 1) it is said that Greenland ["Viridis terra"] nearly touches the African Islands (i.e., Insulæ Fortunatæ), then we possibly have an explanation of this juxtaposition. But as it is said in the same passage that Greenland forms the western end of Europe, we cannot suppose that the cartographer was acquainted with this work. The probability is, no doubt, that Greenland [Illa verde] together with Brazil or the Insulæ Fortunatæ had become transformed into mythical islands out in the ocean.

On another compass-chart, bound up in a Paris MS. of Ptolemy of the latter part of the fifteenth century, a similar island (or peninsula ?), with the same round island to the south of it, is seen to project southwards from the northern border of the chart out into the Atlantic, and a little farther east than the Insulæ Fortunatæ. On the island is written: "Insula uiridis, de qua fit mentio in geographia" [the green island, of which mention is made in the geography].[272] We do not know what geographical work may here be meant; Björnbo suggests that it might be the lost work of Nicholas of Lynn, who again may have used the Historia Norwegiæ. It is striking that the island, besides being connected with a round island like Brazil, but without a name, is placed on this map near the Insulæ Fortunatæ.

This "green island," which thus is probably a remnant of old Greenland, occurs again in various forms and in various places on many sixteenth-century maps.

[Sidenote: Lascaris's journey to Norway and Iceland, fifteenth century]

It is not surprising that information about the northern lands made its appearance also on the maps of this time, as we know that the North was visited more frequently, and sometimes by eminent southerners, from the year 1248, when the well-known Matthew Paris, who, amongst other things, drew a map of England remarkable for his time, visited Norway. Rather is it strange that the direct knowledge thus obtained did not leave more definite traces. Early in the fifteenth century (some year between 1397 and 1448) a Byzantine, Cananos Lascaris, travelled in the North and wrote about it (in Greek). He mentions amongst other things that in Bergen, the capital of Norway ("Bergen Vagen"), money was not used in trading [this must have been due to scarcity of coin]; but in Stockolmo, the capital of Sweden, they had money of alloyed silver. Bergen had a month of daylight from June 24 to July 25. He also says that he himself went to the land of the Ichthyophagi (fish-eaters), "Islanta," from "Inglenia," and stayed there for twenty-four days. The people were strong and powerfully built, they lived only on fish, and they had a summer day of six months [cf. Lampros, 1881].

[Sidenote: Fifteenth-century maps of the world]

It would take us too far here to attempt a mention of all the fifteenth-century maps which have a different representation of the North; but perhaps some of the mappemundi in wheel-form, which were still current at this time, ought to be referred to. We saw that on Vesconte's map of the world accompanying Marino Sanudo's work the coast-lines of the compass-charts in the Mediterranean, etc., had already been introduced. On the Modena map (p. 231) this has also been carried out as regards the North. In the fifteenth century we have various wheel-maps, of which some seem to be more antiquated. Lo Bianco's round mappamundi, in his atlas of 1436, is connected with the compass-charts of that time. Johannes Leardus's round mappamundi, in many editions of 1448 and earlier,[273] likewise shows a strong affinity to the compass-charts, although there is little detail in the delineation of the North. The same is the case with the anonymous round mappamundi in a codex in the Library of St. Mark at Venice [cf. Kretschmer, 1892, atlas, Pl. III., No. 13], but this map has also points of similarity to Vesconte's mappamundi in Sanudo's work, and, amongst other things, it has the same mountain-chain along the north coast of the continent, and the same form of the Baltic.

The round mappamundi in a MS. of Mela of 1417 at Rheims[274] is, on the whole, of a very antiquated type, but its image of the North seems more modern, and it has the same mountain-chain along the north coast of the continent as Vesconte's map. The "Sallust" map at Geneva, of about 1450, is also antiquated, but its Baltic resembles the compass-charts, and the two mountain ridges, one along the north coast of the continent, the other parallel with it in the interior, strongly recall Vesconte's map of the world. On the other hand, the connection by water between the Baltic and Mæotis (the Sea of Azov) is evidently derived from an earlier age (cf. p. 199). Out in the ocean to the north-west and west of Norway lie four islands. Björnbo supposes [1910, p. 75] that the two more northerly of these may correspond to Adam of Bremen's Greenland and Wineland, but this must be very uncertain.[275]

[Sidenote: Walsperger's map of 1448]

A curious delineation of the North is found on the round mappamundi which was drawn at Constance in 1448 by the Benedictine monk Andreas Walsperger of Salzburg [cf. Kretschmer, 1891a]. The map is in most respects imperfect and antiquated, but shows also more recent, particularly German, influence.

The Mediterranean and the Baltic are disproportionately large, and the mass of land between them has been contracted. There are many mediæval mythical conceptions, and items showing possible influence by Adam of Bremen [cf. Miller, iii. 1895, p. 147]. Thus in northern Asia we have "Cenocephali" and Cannibals ["Andropophagi"], bearded women, Gog, Magog, etc. In Norway we read: "Here demons often show themselves in human shape and render service to men, and they are called trolls." Claudius Clavus also speaks of trolls in Norway. In the northern ocean to the north-west of Norway is written: "In this great sea there is no sailing on account of magnets." This is evidently the widely distributed mediæval myth of the magnet-rock, which attracted all ships with iron in them; in Germany it occurs in the legend of Duke Ernst's wanderings in the Liver Sea, and it is doubtless derived from the Arabian Nights. On the mainland to the north-east of Norway we read that "here under the North Pole the land is uninhabitable on account of the excessive cold which produces a condition of continual frost...." In the extreme north of the ocean, near the Pole, is written: "Hell is in the heart or belly of the earth according to the opinion of the learned."

"Palus meotidis" [the Sea of Azov] is marked as a lake due east of the Baltic. Along the north coast of Europe (and Norway) is indicated a ridge of mountains, somewhat similar to that in the Sanudo-Vesconte maps of the world. The delineation of Denmark ("dacia," with "koppenhan" and "londoma," i.e., Lund), the straight south coast of the Baltic, and a long-shaped island called "Suecia" (with "Stocholm" and "ipsala") on the north, remind us a good deal of Edrisi's map (p. 203), and also somewhat of the Cottoniana (vol. i. p. 183). To the north of the island of Suecia "the very great kingdom of Norway ['Norwegie']" projects to the west as a long peninsula bounding the Baltic, with "brondolch" [Bornholm ?] and "nydrosia metropolis" [the capital Nidaros] as towns on its south coast, and with the land of "Yslandia" [Iceland] and the town of "Pergen" [Bergen] on its extreme promontory.

[Sidenote: The Borgia map, after 1410]

Another peculiar type of the round mappamundi is the so-called Borgia map of the fifteenth century (after 1410). Its representation of Europe, with the Mediterranean on the southern side of the earth's disc, is very imperfect and far removed from reality. The same is the case with its delineation of the North, but curiously enough its Scandinavia, which is different from that of the compass-charts, and in which Skåne forms a peninsula on the south, to the east of Denmark, has a greater resemblance to reality than that of other maps of this time. This map, too, has a chain of mountains along the north coast of the continent, as in the Vesconte maps [see Nordenskiöld, 1897, Pl. XXXIX.].

[Sidenote: Fra Mauro's map, 1458]

The best known fifteenth-century map of the world is that of Fra Mauro (1457-59), which is also drawn in wheel-form and is preserved at Venice. The coast-lines are taken to a great extent from the compass-charts, but a great deal of new matter has been added. As regards Norway, this consists of information from Querini's voyage in 1432, as well as from other sources which are unknown to us; this is indicated by, amongst other things, an inscription on the sea to the north of Russia ["Permia"], which relates that a short time before two Catalan ships had sailed thither [cf. Vangensten, 1910]. On this map the Scandinavian Peninsula has been given a more reasonable extension to the north; but the west coast is very imaginatively supplied with peninsulas and islands, while the ocean outside is full of fabulous islands and contains many legends.

Denmark ["Datia"] has been made into an island (which is also called "Isola islandia"), and the Baltic ["Sinus germanicus"] has been widened into an inland sea with islands. In its northern part is a note that on this sea the use of the compass is unknown [cf. Vangensten, 1910]. Could this inscription be due to a misunderstanding like that on the Walsperger map in the ocean to the north-west of Norway, that it could not be navigated on account of magnets (cf. p. 283)? There is no hint of the name of Greenland on this map; on the other hand, Iceland appears in three or four different places: besides Denmark, as mentioned above, there is in northern Norway or Finland a peninsula named "Islant," "where wicked people dwell, who are not Christians"; also a large island, "Ixilandia," north-west of Ireland, and finally an intricate peninsula in the middle of Norway called "Isola di giaza" [i.e., the island of ice]. On the north of Norway or Finland a peninsula projects into the Polar Sea with the name of "Scandinabia." The map does not contribute anything new of importance about the North, but points to a few fresh pieces of information about Norway, which are not to be traced in the older compass-charts; thus Bergen comes nearly in its right place on the west coast, and Marstrand appears to the east of Christiania fjord.

[Sidenote: Genoese mappamundi, 1447]

A picture of the North of a wholly different type is given on the elliptical Genoese mappamundi [of 1447 or 1457], which is still more fantastic than any of those hitherto mentioned. The Scandinavian Peninsula has a very long extension to the west, and ends in a promontory projecting northwards. To the north of this Scandinavia there is another fantastic peninsula where Lelewel thinks he can read the name "Grinland," which is probably due to a misunderstanding, since, as pointed out by Björnbo [1910, p. 80], the name cannot be seen on the much-damaged original, or on Ongania's photographic reproduction [Fischer-Ongania, Pl. X.]. Many imaginary islands are scattered about in the sea round these peninsulas.

[Sidenote: Globes of the fifteenth century]

[Sidenote: Behaim's globe, 1492]

Towards the close of the fifteenth century the discovery was made of representing the surface of the earth, with land and sea, on globes. It was evidently the efforts of Toscanelli that led to the general adoption of this mode of representation, which had been used by the Greeks at an early time (cf. vol. i. p. 78); in 1474 he announced that his idea of the western route to India could best be shown on a sphere. Columbus seems to have taken a globe with him on his voyage of 1492, according to his own words in the ship's log. The oldest known terrestrial globe that is preserved was made in 1492 by the German Martin Behaim (born at Nuremberg in 1459).[276] He spent much time in Portugal, and also in the Azores, after making a distinguished marriage with a native of those islands, a sister-in-law of Gaspar Corte-Real's sister. But it was during a visit to his native town (1490-93) that he constructed his globe. The sources of Behaim's representation of the North were principally Nicolaus Germanus's mappamundi in the Ulm editions of Ptolemy, of 1482 and 1486, where Greenland is placed to the north of Norway, and Marco Polo's travels, which speak of the northern regions of Asia. Besides these a name like "tlant Venmarck" (the land of Finmark), for instance, points to a use of the same older authority as in the anonymous letter to Pope Nicholas V., of about 1450, where in the existing French translation there is mention of "lieux champestres de Venmarche" [the plains of Finmark].[277] Thus we are here again led to the lost work of Nicholas of Lynn, "Inventio fortunata" (1360), as the possible source. That it really was this work that was used seems also to result from the fact that the countries about the North Pole on Behaim's globe bear a remarkable resemblance to Ruysch's map of 1508, where this note is given at the North Pole:

"In the book 'De Inventione fortunata' it may be read that there is high mountain of magnetic stone, 33 German miles in circumference. This is surrounded by the flowing 'mare sugenum,' which pours out water like a vessel through openings below. Around it are four islands, of which two are inhabited. Extensive desolate mountains surround these islands for 24 days' journey, where there is no human habitation."

What is new in Behaim's picture of the North is chiefly this circle of land and islands around the North Pole, which he evidently took from Nicholas of Lynn, and which is not represented on any older map known to us. It consists of a continuous mass of land proceeding from his Greenland-Lapland to the north of Scandinavia, and extending eastward nearly to the opposite side of the Pole, where the Arctic Ocean ("das gefroren mer septentrional") to the north of the continent becomes an enclosed sea. On the other side of the Pole are two large islands and a number of smaller ones. On one of the large islands is a picture of an archer in a long dress attacking a polar bear (which may be connected with myths about Amazons ?), and on the other side is written: "Hie fecht man weisen valken" [here they catch white falcons]. It might be supposed that this was derived from statements about Scandinavia or Iceland (cf. e.g., the legends of the compass-charts); but, as assumed by Ravenstein [1908, p. 92] and Björnbo [1910, p. 156], it is more likely to come from Marco Polo's travels, where the Arctic coast of Siberia is spoken of. The many correct names, in a German form, in Martin Behaim's Scandinavian North point to the possibility of his also having received oral information, though they may equally well be derived from older German maps.

[Sidenote: Laon globe, 1493]

Almost contemporary with Behaim's globe is the so-called Laon globe of 1493, which was accidentally discovered in a curiosity shop at Laon some years ago. It gives a wholly different representation of the North, more in agreement with the usual maps of the world of the Nicolaus Germanus type, with sea at the pole round the north of the continent, which terminates approximately at the Arctic Circle. The Scandinavian Peninsula (called "Norvegia") has a form somewhat resembling this type; but to the north of it "Gronlandia" appears as an island, with a land called Livonia projecting northward on the east, and two islands, Yslandia and Tile, on the west. Nothing is known of the origin of the Laon globe, or of the sources of its representation of the North.

Such were the geographical ideas of the North at the close of the Middle Ages, when the period of the great discoveries was at hand; they were vague and obscure, and the mists had settled once more over large regions which had been formerly known; but out in the mists lay mythical islands and countries in the north and west.