But the Florentine map of 1351, called the Laurentian Portolano, is to all appearance a record of the actual discoveries of 1341 and 1346, and a wonderful triumph of guess-work if it is nothing better. For Africa is not only made an island, but the main outline of its coast is fairly drawn; in its western corner the headlands, bays, and rivers are laid down as far as Bojador, and the three groups of Atlantic islands, Azores, Canaries, and Madeira, appear together for the first time. Beyond this names grow scarce, and on the great indent of the Gulf of Guinea, enormously exaggerated as it is, there is nothing to show for certain any past discovery, which suggests that this map was made for two purposes. First, to record the results of recent travel; secondly, and chiefly, to put forward geographical theories based upon tradition and inference, what men of old had told and what men of the present could fancy.
Long after the Italian leadership in exploration had passed westward, Italian science kept control of geographical theory; the Venetian maps of the brothers Pizzigani in 1367, and of the Camaldolese convent at Murano in 1380 and 1459, and the work of Andrea Bianco in 1436 and 1448, are the most important of mediæval charts, after the Laurentian, and along with these must be reckoned that mentioned above as given in 1425-8 to Henry's brother, Don Pedro, on his visit to Venice. This treasure has disappeared, but it was said by men of Henry's day and aftertime, who saw it in the monastery of Alçobaça, to show "as much or more discovered in time past than now." If their account is even an approach to the truth, it was in itself proof sufficient of the supremacy and almost monopoly of Italians in geographical theory.
With 1375 and the Catalan map of that year, which specially refers to the Catalan voyage of 1346 and may be taken as one result of the same, we come to Spanish parallels; but until the death of Henry in 1460, Italian draughtsmen were in possession, and Fra Mauro's great map of 1459, the evidence and result, in great measure, of the Navigator's work, could only be drawn by Venetians for the men whose discoveries it recorded.
But there is one other point in Italian map-science which is worth remembering. At a time when most schemes of the world were covered with monsters and legends, when cartography was half mythical and half miscalculated, the coasting voyagers of the Mediterranean had brought their Portolani or sea charts to a very different result. And how was this? Did they get right, as it were, by chance? "They never had for their object," says the great Swedish explorer and draughtsman, Baron Nordenskjold, "to illustrate the ideas of some classical author, of some learned prelate, or the legends and dreams of feats of Chivalry within the Court circle of some more or less lettered feudal lord." They were simply guides to mariners and merchants in the Mediterranean seaports; they were seldom drawn by learned men, and small enough, in return, was the attention given them by the learned geographers, the men of theory, in the fifteenth and sixteenth centuries.
But these plans of practical seamen are a wonderful contrast in their almost present-day accuracy to the results of theory let loose, as we see them in Ptolemy and the Arabian geographers, and in such fantastics as the Hereford Mappa Mundi, so well known in England. Map-sketches of this sort, were unknown to Greeks and Romans, as far as we can tell. The old Peripli were sailing directions, not drawn but written, and the only Arabian coast-chart known to us was copied from an Italian one. But from the opening of the twelfth century, if not before, the western Mediterranean was known to Christian seamen—to those at least concerned in the trade and intercourse of the great inland sea,—by the help of these practical guides.
THE LAURENTIAN PORTOLANO OF 1351.
[(see list of maps)]
From the middle of the thirteenth century, when the use of the compass began on the coasts of southern Europe, the Portolani began to be drawn with its aid, and by the end of the same century, by the time of our Hereford map (c. 1300), these charts had reached the finish that we see and admire in those left to us from the fourteenth century. For, of the 498 specimens of this kind of practical map now left to us, there is not one of earlier date than the year 1311. Among these specimens not merely the mass of materials, but the most important examples, not merely 413 out of 498, but all the more famous and perfect of the 498 are Italian. The course begins with Vesconte's chart, of the year 1311, and with Dulcert's of 1339, and the outlines of these two are faithfully reproduced, for instance, in the great Dutch map of the Barentszoons (c. 1594), for the type once fixed in the fourteenth century, recurs steadily throughout the fifteenth, and sixteenth. The type was so permanent because it was so reliable; every part of the Mediterranean coast was sketched without serious mistake or disproportion, even from a modern point of view, while the fulness and detail of the work gave everything that was wanted by practical seamen. Of course this detail was in the coast lines, river mouths, and promontories; it only touched the land features as they touched the seas. For the Portolani were never meant to be more than mariners' charts, and became less and less trustworthy if they tried to fill up the inland spaces usually left blank. For this, we must look to the highest class of mediæval theoretical maps, those founded on Portolani, but taking into their view land as well as water and coast line. And such were the celebrated examples[29] we have noticed already.