Ancient maps were not without high merits in certain cases, and a little after Prince Henry's time the Renaissance editions of Ptolemy played a very important part in geographical history. But in the first part of the fifteenth century neither the work of the Alexandrian astronomer and cartographer, nor the ancient road maps of the Roman Empire and surrounding lands[[236]] seem to have been sufficiently known for the exercise of much influence in the progress of discovery or of geographical knowledge. The same result follows, for different reasons, in the case of almost all the earlier mediæval maps and charts,[[237]] which are quite unscientific in character, and often rather picture books of natural history legends than delineations of the world.
Strictly scientific map-making begins with the Mediterranean portolani. The earliest existing specimen of these is of about 1300, but the type then formed[[238]] must have been for some time in process of elaboration; and it is even probable that a fully-developed example from the middle of the thirteenth century may yet be discovered.
"A sea-chart—probably a portolano—is mentioned as early as the account of the Crusade of St. Louis, in 1270."[[239]] So in Raymond Lulli's Arbor Scientiæ, written about 1300, we have reference to compass, chart and needle, as necessary for sailors.[[240]] Once again, it is probable that Andrea Bianco's planisphere of 1436[[241]] is only a re-edition of a thirteenth-century work, when the "Normal Portolano" was just in process of making, but had not reached even the comparative perfection of the Carte Pisane, Carignano, or Vesconte examples.
The earliest dated portolan is that of 1311, by Petrus Vesconte; and from this time the maps of this class, whose central feature is an accurate Mediterranean coast-line, increase rapidly, being indeed all reproductions of one type,[[242]] occasionally introducing additions or corrections, especially in outlying parts, but not often varying much from one another in the central portions. The type is reasonably believed by some[[243]] to have originated among the Catalans, either of Spain, France, or the Balearic Isles, well within the thirteenth century.[[244]] In connection with this, we may recall the point mentioned by Barros, that Prince Henry the Navigator obtained the services of Master Jacome, or James, from Majorca to instruct the Portuguese captains in navigation, map-making, and the proper handling of nautical instruments.
These plans of practical seamen are a striking contrast, in their often modern accuracy, to the results of the literary or theological geography portrayed in such works as those of the "Beatus School," or of Robert of Haldingham.[[245]] Map surveys of this kind were apparently unknown to the ancient world. The old Peripli were sailing directions, not drawn but written; and the only Arabic portolan known to exist was copied from an Italian example. Long after the Italian leadership in exploration and commerce had begun to pass away, Italian science kept control of cartographical work; thus, among the early portolani, not only the majority—413 out of 498—but the most valuable, were executed by the countrymen of Carignano and Vesconte.
This department of geographical history is only just beginning to be appreciated at its full value—as marking the vital transition from ancient to modern, from empirical to scientific—but this need not surprise us much. The portolani, as has been well said, never had for their object to provide a popular or fashionable amusement; they were not drawn to illustrate the works of classical authors or learned prelates; still less did they illustrate the legends and dreams of chivalry and historical romance; they were seldom drawn by learned men; and small enough in return was the acknowledgment which the learned but too often made them, when the great geographical compilers of the Renaissance and Reformation times incorporated the earlier coast-charts in grander and more ambitious works.
Unquestionably, however, it is in maps of the portolano type that we must look for Prince Henry's primary geographical teachers, though the influence of books—and even of the older theoretical designs in cartography—must not be forgotten. Therefore, to understand his position—to realise what he had to draw from—we must briefly describe the chief designs which it was possible for him to consult for his scientific purposes, for his Ptolemaic ambition, διοθῶσαι τὸν ἀρχαῖον πίνακα.
(1) The "Carte Pisane" of the latest thirteenth or earliest fourteenth century is probably only a copy of an earlier work, though now itself our earliest example of the portolano type. The Mediterranean on this example (as well as the Black Sea, where it has survived injury) shows the new scientific or surveying method, but the Atlantic coasts of Spain and France, and still more the shore-lines of Britain, are of a different and inferior character. This alone points to an earlier date than, e.g., the works of Vesconte and Dulcert. In West Africa only a part of the Maroccan coast now remains.
(2) The Map of Giovanni di Carignano,[[246]] of c. 1300?-1310, though much damaged, shows the Black Sea and Britain with contours differing somewhat from the ordinary portolan; and the same is noticeable in the Baltic. The West African coast does not extend to Cape Non. Another work by Carignano, of c. 1306, "specially referring to Central Asia," is said to exist, but its present position is unknown.
(3) A portolan of the early fourteenth(?) century, belonging to Professor Tammar Luxoro, of Genoa, in 1882, and usually called after him, is believed by Nordenskjöld to be a "slightly altered copy of the normal portolano in its original form." In N.W. Africa it only gives us the shore-line as far as Sallé, with a series of names, beginning at Arzilla.[[247]]