WWW-Virtual Library: History
Map History / History of Cartography: THE Gateway to the Subject

WWW-VL Main Catalogue WWW-VL History Central Catalogue
(main menu)


What the
site is
Map History

Cartographic innovations by the early portolan chartmakers

(and subsequent developments)

focusing through the centuries on the traditional medieval coverage:
the Mediterranean and Black Seas, the North Sea and Baltic, and north-west Africa 1

by Tony Campbell

Copyright © 2016-2024

Mounted on the web 18 November 2016 – with subsequent corrections and additions noted thus {...December 2020}, which can be searched for

The aim of this online Note is to bring together, in a summary form, scattered facts from portolan chart history.
Because of the limited documentary evidence, much of what follows has had to rely on close observation of the charts themselves


To get from the note number to the endnote text, and vice versa, just click on the number

For the current locations of the charts referred to in this Note see List of the charts consulted

Introductory notes on what are considered to be the earliest sources

Portolan chart origins

Before any cartographic innovations can be confidently attributed to the portolan chartmakers, agreement is needed as to when those documents originated. The ‘First International Workshop on the origin and evolution of portolan charts’ (Lisbon, 6-7 June 2016) reached a broad (though not unanimous) consensus around a High Medieval origin, even if detailed evidence remains elusive. 2   A date early in the 13th century (or conceivably a little before that) may be indicated by the apparent reference to a prototype marine chart [cartula mappe mundi] in a unique manuscript text. 3   However, if that is the case, no further textual references have been noted until many decades later, in the 1270s, in other words around the likely time of the oldest known chart. For a detailed investigation into the origin of the portolan chart, concluding that their first appearance was likely to have occurred around 1200, see ‘Mediterranean portolan charts: their origin in the mental maps of medieval sailors, their function and their early development’ {published online on 26 January 2021 }.

Dating the Carte Pisane

The earliest surviving chart, the Carte Pisane, one of the great treasures of the Bibliothèque nationale de France, had been assigned by most researchers to the end of the 13th century. Then, in 2012, at an international conference held in that institution, Ramon Pujades claimed that it (and two other, evidently closely related, unsigned charts) should be moved forward perhaps even a century or more. 4   Those arguments were challenged by the present writer in a detailed online essay. 5   To resolve this issue, which was seriously disrupting the discussion about the origin and earliest history of the portolan chart, the BnF commissioned a carbon-dating analysis. At the Lisbon conference already mentioned, Catherine Hofmann announced that the vellum used for the Carte Pisane had been C-14 dated to 1170-1270 (with a stated probability of about 95%). 6

If that result provided confirmation that the Carte Pisane was indeed the oldest surviving portolan chart, it still left unresolved that work’s most plausible date. Much of the 100-year window of the C-14 dating had to be ignored because of the chart’s own geo-political chronology. Inclusion of various known foundations, for example Manfredonia (1258), Claren[t]za (1260s) and Gioia Tauro (1271) – in each case clearly an original feature of the chart, not a later addition – forces a terminus post quem at the very end of the C-14 range. 7

Additionally, the inclusion of Palamos, founded in 1279, had seemed to push on the latest possible date for the Carte Pisane, slightly further again. However, it has been recently suggested that the new settlement might have taken its name from the curved back of that bay, which could have been what the Carte Pisane was referring to instead. 8

The C-14 analysis concerned itself only with the date of the animal skin, not with the chart drawn on top of it. Even though parallel testing of the ink and pigments found nothing to doubt the 13th-century date, nor did it add new evidence. Nevertheless, it is theoretically possible that the skin was prepared and then left, for an unknown period, before being used for a marine chart. Possible, but surely unlikely, given that once money had been expended on a prepared skin it would have made sense to use it without delay. If a ‘delayed use’ argument is to be deployed it would be necessary to answer two questions: first, is it easier or more difficult to draw on a piece of vellum that had been stored (and hardened) for, perhaps, decades; and, second, is such a practice recorded for that period?

A further point relates to the question of the Carte Pisane’s primacy. It has been generally considered that a chart now in Cortona, described by Vera Armignacco in 1957, while placed earlier than the first dated chart (Pietro Vesconte’s of 1311) was still produced later than the Carte Pisane. 9 Again at that Lisbon workshop, an expert in the history of southern France and its Mediterranean coastline offered evidence to suggest that the Cortona chart may even predate the Carte Pisane. 10   That contention would need to be tested for the entire portolan chart coverage, and my own more general analyses had not led me to doubt the Carte Pisane’s primacy. But, if that suggestion was confirmed, it would probably involve moving the Cortona chart to a date a little before that now occupied by the Carte Pisane, rather than forcing the movement of the latter even further beyond the C-14’s latest indicated date.

It is highly unlikely that the Carte Pisane – a rare, chance survivor – with its already fully recognisable outlines for the Mediterranean and Black Seas, can give us much idea of the appearance of the lost prototype charts. But it (and the Cortona chart) provide us with the only cartographic evidence from what was still a period of significant development. So, while giving due credit to the ‘Liber de existencia riveriarum’ (and perhaps to the other 13th-century portolano, ‘Lo compasso de navegare’), 11 for providing full itineraries of the Mediterranean coast’s ports and natural features, while simultaneously stating the distance and direction from one to the next, those are textual not cartographic works and may have served a related, but somewhat different purpose. In brief, a chart was of far more practical utility for planning a voyage and then more convenient for use when on board ship.

General prefatory remarks

Since the portolan charts were the earliest systematic and dedicated cartographic aids for marine navigation, it is inevitable that various hydrographic features will be appearing there for the first time. But, beyond that, the inventiveness of successive chartmakers led to the introduction of a number of conventions into cartography as a whole. Indeed, it is not unreasonable to claim that those charts, in the form they had achieved by the end of the developmental period (around 1330), embodied more cartographic innovations than any other map type. The Portuguese historian Armando Cortesão considered the portolan charts to represent one of the most important turning points in the history of cartography. 12 Alberto Magnaghi went further, describing them as a unique achievement not only in the history of navigation but in the history of civilization itself. 13   It is hoped the observations which follow will substantiate those statements, at least to some extent.

This Note, and indeed the entire suite of articles that make up the portolan chart section of the ‘Map History’ site, 14 would not have been possible without the dedication of Ramon Pujades in gathering onto a single DVD almost all of the charts and atlases produced before 1470. The quality of those scans, the ability to enlarge them to a high level of visibility, and, most important for toponymy, the rotation option, finally opened up research in this field. For the first time, those with access to that invaluable 2007 resource are presented with the essential raw material of our subject (if only via reproduction). 15

The following claims for innovations within the genre of marine charts are unlikely to be challenged on factual grounds, given the availability of that Pujades DVD. However, assertions that some features pre-empt cartography in general may be open to question. Comments are invited on the matters discussed here. However, speculation about portolan chart origins and findings based on cartometric analysis would need to be addressed elsewhere, to those more able and willing to respond.

Please send corrections and additions to the author Tony Campbell:  

Top of page

The Innovations

Present from the outset

The largest single innovation is undoubtedly an idea: the concept of a portolan chart. Also the realisation that an animal skin, untrimmed, could be used to carry a map of the Mediterranean (is there any evidence that entire skins had been used by scribes in that way before?) {this sentence added 14 May 2019}. With no precursors identified for any of the specific instances described below, where could this wide range of cartographic devices have come from? That question touches on the unresolved issue of portolan chart origins. Roel Nicolai has recently claimed that the Middle Ages had neither sufficient knowledge nor adequate technology to have produced those charts, which must, somehow, have been passed down in their completed form from an earlier, unspecified period {corrected to "passed down as partial charts", 8 January 2017}. 16  If the prodigious display of cartographic novelties on the earliest surviving chart, the Carte Pisane, might be able to co-exist with that argument, the major hydrographic improvements and the significant inventiveness observed in the work of chartmakers during the few decades that followed evidently contradict it. Lacking any recognised antecedents, the essentially practical purposes behind those innovations point to their most likely origin within a distinctly maritime milieu, presumably an active Italian port where ideas and prototype charts could have been shared, commented upon and improved.

At this stage we cannot go further with any confidence. In time, the growing sophistication of cartometric methods may lead to a single understanding of the charts’ underlying geometry, around which an explanation of how they were constructed could coalesce, both with the limited historical data available to us and the evidence drawn from the documents themselves.

These Notes strive to be purely factual, avoiding the copious speculation that continues to surround these mysterious documents. I hope the following brief exception to that rule will be excused.

It seems possible (as a thesis which I hope to elaborate later) 16a  that the concentration by cartographic historians on the mathematical underpinnings, such as projection, measured angles, scaled distances, and so on, may be obscuring a different process. For millennia, men had sailed, not just around but across the Mediterranean, apparently without the aid of a map. Except for the one they must have held in their head. A professional sailor involved in a variety of voyages over decades would inevitably have built up a mental chart, even if unconsciously. After all, before he set off on any voyage he would have needed to know how far to sail (or row), in what direction, and with what intervening landfalls. Such a ‘chart’, however widespread these might have been, would not have taken tangible form until someone decided to sketch one out in ink on an animal skin.

This mental construct could not have derived from existing maps. How could it have been built on the cartography available in the 12th-century, comprised of T-O diagrams, the largely schematic or stylised mappaemundi, on itineraries or local maps, etc.? 17   Nor would the maps of Ptolemy have yet been available to the portolan chart’s creator(s). Should we be so surprised that its content was expressed in novel ways?

The new conventions highlight the fact that the portolan charts’ creators could never have seen a map before, or at least not one that proved useful to them. In devising a pragmatic working tool they therefore started from first principles. The word ‘invention’ seems appropriate, given that a number of the chart’s conventions could, in a modern age, have been patented, since they were new and involved inventive steps {this paragraph added 14 May 2019}.

Without any help from precedent, the portolan chart was born to serve a number of specific, highly practical navigational purposes. The impetus must have come from below, not above. Originality was therefore inevitable.

The following is a suggested list of the achievements of the earliest portolan charts:

  • along with the magnetic compass, the hour-glass, the lead-line and the marteloio, 18 they are the earliest artificial aids to navigation in the West
  • what they offered seamen was unparalleled in its complexity and practical usefulness compared to anything made for landsmen 19
  • they are the first maps of the Mediterranean and Black Sea that are fully recognisable to us 20
  • the earliest surviving charts already provide (for the first time in cartography) clear, complete and realistic coastal outlines, placed within a broad context that included all the seas in which medieval Mediterranean mariners might expect to sail. Approximate spatial accuracy, the avoidance of significant, possibly fatal errors, was vital to their purpose. In that respect they differ from all smaller-scale maps of the time
  • the charts position the headlands in the Mediterranean and Black Seas in relation to their neighbours (broadly correctly) as well as to those on the opposite shores
  • they are the first maps to give clearly recognisable shape to the linked gulfs of Gabès and Sirte (Sidra). Extending across much of Tunisia and Libya, this major indentation into the African coastline (which was no more than hinted at in the Ptolemaic maps) is the most visible signature of the portolan charts. Its presence on a world map is evidence of a clear debt to the charts, even if indirectly 21
  • the portolan charts were concerned exclusively with the coastlines but as an accidental result they thereby delineated the outer limits of the countries involved. {14 May 2019}
  • the charts’ toponymy is dense (the Carte Pisane already has over 600 names for the Mediterranean coastlines). It also provides a contemporary record of name forms then in current use among mariners – since only a few refer to classical sites – if allowance is made for variations in language and dialect
  • the range of (often unstated) referents for the toponyms is probably unprecedented, with a mix of coastal, offshore or inland features. This constructive ambiguity enabled the charts’ user to attach whatever referent they liked to the name – perhaps more than one. {14 May 2019}
  • unlike contemporary maps, the portolan charts do not use dots to locate places. This allowed the toponyms to be spread out evenly along a coastal stretch, thus saving the copyist’s time. {14 May 2019}
  • among drafting conventions not apparently seen previously is the placement of the littoral toponyms perpendicular to the coast and inland, in a single sequence whose changing orientation would have required the chart to be rotated when in use, in order for the place-names to be readily legible 21a
  • red ink (a well-known scriptorium device) is used to distinguish toponyms of greater significance (about one-sixth of the total) from the remainder in black, whereas other medieval maps reserved red for regional names. The judgements as to which merited red were made individually, by many different chartmakers over the centuries. An estimated 90% of the red names refer to ports rather than headlands or river mouths 22
  • by stripping the seas of their habitual colour or patterning, this removes any distraction from the charts' maritime focus 23
  • there is a systematic network of 16 / 32 wind or compass directions (rather than the 8 or 12 of medieval geographers) based around one or two large hidden circles. This is unconnected to the coastal outlines, though they have a common orientation
  • the earliest surviving charts contain the first indicated scale on a map: one or more of its varying number of larger divisions would be subdivided into five sections, each representing ten miglia. Because the value was not stated, this remains uncertain. [One might conjecture that the user would have been happy to reverse the usual process and derive the figure from his own experience of local distances, or the value was conveyed orally.] Initially, three different scales were used (for the Atlantic, Mediterranean and Black Sea respectively) – if not for the individual basins as well. Until 1318, when the rectangular bar is first seen, the scale was arranged in various ways within a circle 24
  • they introduced a measurement, the ‘portolan mile’, which whatever its length (and the value for the Atlantic, Aegean and Black Sea apparently differed), would presumably have become standardised through use of the ubiquitous charts {7 July 2020}
  • navigational dangers, which may indeed have provided one of the main reasons for the charts' invention, are noted in simplistic form on the Carte Pisane. By around 1400, a new and unique symbolic language had been elaborated into a complex coded system of crosses and dots, in black or red, to distinguish rocks and reefs, sandbanks and shoals. 25 Those features, among them the much-feared Skerki Bank west of Sicily, will presumably have made their first cartographic appearance on the portolan charts, and some can be recognised on their successors today. Whereas 'Lo compasso' does provide useful warning notes, being able to locate and depict the offshore hazards highlights one of the portolan charts’advantages over the textual portolani
  • they can lay claim to the title of the first transport or mobility map – the first wayfinding map designed to be used outdoors, to guide travel rather than just offer information – to be followed by those which traced the routes of canals (from the 14th century in Europe) {7 July 2020 & 25 February 2021}
  • before printing, they were the most widely distributed documents apart from religious texts {25 February 2021}

Top of page

Can a ‘portolan chart’ be defined or merely recognised from its characteristics?

No adequate definition of a portolan chart has yet been devised. In the opinion of a recent author, ‘If a chart has been constructed in a certain way and if its appearance fits the established criteria, for our purposes, it is a portolan chart’. 26  My own failed attempt at a definition concluded that it was not possible to go beyond ‘a list of stylistic characteristics’. 27  In other words, we are saying no more than that a portolan chart is something that looks like a portolan chart! (but see below under Geometric Developments)

Leaving aside the penultimate entry below (a list of illustrations, which would not have been included on working charts anyway) the apprenticeship process, through which most professional practitioners must have passed, would have ensured the inclusion of almost all the elements listed below. Hence we would expect to see most of the following characteristics on what might justifiably be termed a ‘portolan’ chart, namely that it:

  • is hand-drawn on vellum, usually on a single prepared animal skin or as a bound collection of sectioned charts
  • uses black and red ink, along with various watercolour tints (conveying specific meanings not just adornment)
  • has a systematic 32-line network of compass directions (colour-coded in black, red and green), which is formed around one or two ‘hidden’’ circles’, usually of the maximum height allowed by the vellum
  • has the mainland place-names (in black, or red for the more important places) written perpendicular to the coastline, and placed inland so as to leave the sea clear (which is itself free of colour or patterns)
  • usually denotes rivers on Italian charts by two short parallel lines marking the mouth [apparently another portolan chart innovation, 26 August 2018], though Catalan charts include river courses among the inland detail
  • often conveys the confusing nature of the major estuaries by means of symbolic islands, differently coloured and artificially shaped
  • uses coastal generalisation in the form of pointed or regularly sculpted headlands, with evenly scooped bays between them
  • displays (from about 1330 to 1600) a series of generally repeated, artificial 'signature' shapes for islands, particularly in the Aegean Sea
  • provides sparse geographical information about the interior
  • includes representations of navigational dangers (rocks and reefs, sandbanks and shoals) in a complex symbolism of red and black dots and crosses
  • incorporates (most noticeably on Catalan-style charts and more generally from the 16th century) illustrative details, such as city flags, simplified town symbols or views, images of rulers (with their thrones or tents), regional fauna, green-coloured mountain ranges, the appropriately-coloured Red Sea, compass roses, ships, religious images, wind-blowing putti , etc., in various styles, which can aid in attributions of authorship (or at least the regional school) and dating. 28
  • lacks any key or explanation, despite being partly textual, so that the length of a mile, the meaning of the danger symbols, the identity of the city emblems, conventions such as that for the river entrances, etc., could not have been understood without oral exegesis 29

Whether or not all the features described above are considered essential for inclusion within the genre, all can still be found on such charts over the next four hundred years, i.e. up to around 1700. 30  Such longevity, for a largely unchanging format that was retained for its practical utility, may well be unprecedented in cartography

Top of page

Innovations first noted in 1311 and those added later

(What follows is a mixture of significant observations and a few minor details that may help with chart dating. It is assumed here that the Riccardiana chart is later than 1311, though that may not be the case)

1311-c.1330 (Pietro and/or Perrino Vesconte)

  • Pietro Vesconte's earliest surviving work, his chart of 1311, provides the first dated instance of one of the charts’ most distinctive characteristics, the use of three alternating colours to distinguish the 32 ‘wind’ or compass directions: eight primary in black, eight secondary in green and 16 tertiary in red
  • the 1311 chart is also the first to employ three colours to differentiate neighbouring islands. For his 1313 atlas (the oldest to survive) Vesconte used seven tints, but the average on the charts over the next two centuries was about six 31
  • unprecedented mnemonic island shapes were introduced, particularly for the highly complex Aegean archipelago, by Vesconte in 1311 and then extended by Dalorto / Dulceti in 1330 – another instance of the chartmakers' inventiveness in finding a clever solution to a practical problem 32
  • Vesconte (or perhaps Marino Sanudo), introduces a range of town flags (some new to vexillology) c.1320, unless the Lucca chart’s array preceded that 33

  • Vesconte's charts mark the beginning of systematically updated coastal toponymy (1311-c.30) 34
  • almost all the early charts that have a full geographical coverage use a twin network of compass directions. This meant that the coasts within the triangular spandrels either side of the point of tangency where the two main circles meet were not automatically provided with compass lines (especially over the Adriatic Sea). Various geometrical solutions to that problem were devised. Some may pre-date the first dated instance, Perrino Vesconte’s chart of 1327: for example, the Riccardiana and Lucca charts. The first charts to provide a single network, with extensions to west and east, were produced in Majorca in the late 14th century (Pujades 2007, DVD: C 14 & 15). Few twin networks are seen after 1420.
  • Vesconte (1313) produced the oldest surviving bound collection of sectioned charts, which was to remain a largely Venetian convention
  • Vesconte (1318) is the first to take the scale out of a circle and place it in a rectangular scale bar, having introduced a bi-directional scale in 1311 35

(By the time of the latest Vescontian works, c.1325-30, the coastal outlines of the Mediterranean and Black seas had effectively reached their final forms. There was evidently insufficient demand from later users for hydrographic improvement; indeed, most subsequent changes, particularly from the mid-16th century, represent deterioration)

1330-c.1340 (Angelino Dalorto / Dulceti)

  • the Dalorto / Dulceti chart of 1330 introduces several of the illustrative elements associated with Catalan work: green for prominent mountain ranges (most noticeably the Atlas Mountains), the appropriate colour for the Red Sea, a white cross over Rhodes (in honour of the Knights Hospitaller), ethnographic illustrations, and small stylised diagrams for towns. There were also discursive texts (legends). The 1339 chart (perhaps because of its more luxurious nature and an evident focus on display rather than nautical use) includes portraits of enthroned rulers in addition 36
  • a compass rose (a circular diagram highlighting the eight prominent directions) is seen in 1330 for the first time. Later developments include the Catalan Atlas’s version (c.1375), which shows the full 32 directions. In both those cases the rose was separate from the compass network. The first chart to place the compass rose over one of the points on the periphery of the hidden circle (where the 32 compass lines intersect) is probably the Cresques atelier chart of the late 14th century (Pujades 2007, DVD: C 19), or that by Sentuzo Pongeto (1404). The jointly-authored Corbitis and Pinelli-Walckenaer atlases (probably from that same period), place their 16-point compass roses at the centre of the network instead. The addition of a north-pointing fleur-de-lis outside the compass disk is first seen on the 1492 chart by the Portuguese, Jorge de Aguiar. From the 16th century onwards, compass roses increase in both quantity and intricacy 37
  • the 1330 chart also pioneered the convention of leaving uncoloured rectangular panels (labels) within a coloured island into which its name could be written 38

1367 (the Pizzigani brothers)

  • realistic town views (of Venice and Genoa) appear here for the first time, although the 1327 Perrino Vesconte chart had introduced symbolic fortress images for cities, and a few of the urban symbols in the undated Lucca chart include recognisable elements 39
  • the first illustrations of ships are seen here (in the Atlantic and Indian Ocean) 40

1375 (the Catalan Atlas)

  • the red and gold bands of Aragon are first noted, running diagonally across its creator’s home, the island of Majorca

1403 (Francesco Beccari)

  • in his chart of that date Beccari explains that he had corrected the previous understatement of the scale used in the Atlantic, when compared with that for the Mediterranean 41

1456 (Petrus Roselli)

  • between 1449 and 1456 Roselli “decided to double the number of spokes radiating from the main [compass] center by adding a further sixteen red lines between the existing ones. These run out to the edge of the chart, avoiding all the secondary centers”. His later charts repeat that. 42

1464 (Petrus Roselli)

  • places the first surviving religious image on the neck of a chart. The Madonna and Child or a Crucifixion scene would commonly appear there afterwards.

16th century

  • driven out of Rhodes by the Ottomans in 1522, the Knights Hospitaller were given Malta in 1530. Their red cross appears soon afterwards though most charts retain that for Rhodes as well, even as late as the 17th century

17th century (or very late 16th)

  • two features, which start to appear about 1600, seem to have been English innovations, though further research is needed here. Certainly they were popularised by the Thames-side chartmakers and became standard elements of their productions:

Since my own work has concentrated on the period up to 1500, other 16th- or even 17th-century examples may need to be added to the above)

Top of page

To get from the note number to the endnote text, and vice versa, just click on the number

For the current locations of the charts referred to in this Note see List of the charts consulted


Other developments

Extending the geographical coverage up to 1500


To the North and West


Rudimentary suggestions of the Scandinavian landmass appear before c.1320, when world maps by Paulinus Minorita (Fra Paolino) and Pietro Vesconte included simple, but different indications of the Baltic Sea. Giovanni Carignano’s lost world map / chart followed soon afterwards. Variations of the more detailed outline, first seen on the Dalorto / Dulceti chart of 1330, would follow throughout that century and the next, with little noticeable improvement apart from an added hint of the Gulf of Bothnia. Portolan charts that were extended to cover the known world, such as the Catalan-Estense world map, provide occasional exceptions. 45

North Sea:

The highly simplified outlines given to the continental North Sea coasts on the 1313 Vesconte atlas, along with a toponymy that steadily increased in his output over the next two decades, would remain the standard portolan chart pattern, almost unchanged until the 17th century 46

British Isles: 47

  • The first recognisable south coast of England can be seen on Vesconte’s atlas of 1313

  • The 1313 atlas had omitted Scotland, but the two Vesconte atlases of 1318 show a channel separating England from Scotland (apart from a 'hinge' at the mid-point). This convention would be revived (if only faintly) by Francesco Beccari (1403) and appears regularly thereafter

  • Ireland and the Isle of Man appear first on Vesconte works of c.1320-1

  • Two large mythical (or misunderstood) islands appear off Scotland on the 1330 Dalorto / Dulceti chart: till (north east) and scurce (south-west) 48
  • Two mythical islands are inserted south-west of Ireland: montorio / brazil (first seen on the 1330 Dalorto / Dulceti chart), and man (first seen on the Pizzigani brothers’ chart of 1367 – unrelated to the genuine Isle of Man off north-east Ireland) 49

Atlantic islands:

  • Canary Islands: Lanzarote (with its red cross), Lobos and Fuerteventura (Dulceti 1339); the remaining islands added by the Pizzigani brothers in 1367

  • Madeira (insula caprara) and its two small neighbours (Dulceti 1339)

  • Azores: first documented in 1427 but some commentators suggest that representations of them may be present on 14th-century charts

  • Cape Verde Islands: eight islands (Benincasa 1468); increased to eleven islands (Benincasa 1469) 50
  • Various mythical or misplaced ‘islands’ are shown in the Atlantic, among them brasil and antilia / antillia, prompting speculation that those represent previous but unrecorded American discoveries 51

Western Africa:

The coast appears as far as Mogador / Essaouira on Vesconte’s 1318 atlas (the version in Venice). Thereafter the relationship between features that were first reported, and later actually observed, becomes difficult to disentangle (as also their toponyms). Andrea Bianco 1448, followed by Petrus Roselli and Grazioso Benincasa in the 1460s, include significant southwards additions (both of the coastline and islands), leading up to the discovery of the Cape of Good Hope in 1488. 52


To the East

Some Catalan works, starting with the 1339 chart by Angelino Dulceti, extend to the east to include part or all of Asia. At their greatest extent, for example with the Catalan Atlas (c.1375), these have developed into charts of the entire world, as it was then known or conjectured. While the sections of the Catalan Atlas are presented as a homogeneous whole, a distinction can be seen between the direct experience of the regions covered on the two western sections (a traditional portolan chart) and the theoretical cartography on the two to the east, based on reports or myths. [As pointed out by David Jacoby (personal communication {6 December 2016}), the additional Asian detail must have come from Genoese and/or Venetian sources, since Catalan merchants did not trade in those regions.]

Black Sea:

The section covering the Black Sea was evidently added later to both of the 13th-century portolani: the’ Liber de existencia riveriarum’ and ‘Lo compasso de navegare’ (where the relevant descriptions starts after the colophon). 53   However, it already appears on the earliest surviving portolan charts. The Carte Pisane’s author intentionally omitted the eastern third of the sea, whose included sections are unfortunately of limited legibility (particularly in the chart’s present state). Neither the Carte Pisane nor the Cortona or Riccardiana charts portrays the long west-east Crimean island/sandbar, Tendrovskaya Kosa, nor the Kalamitskiy Zaliv. These were left for Vesconte to include in 1311, as part of the improvements that were to be followed, in the main, by subsequent chartmakers. 54

Caspian Sea:

The Caspian Sea appears occasionally, usually as an eastwards extension on Catalan work, with the earliest example being provided by the Catalan Atlas (c.1375). Cyrus Alai considered the portolan chart’s outline for that sea to be superior to all Ptolemaic and European maps produced before the mid-17th century. 55

(This Note concentrates on the traditional portolan chart areas and hence does not consider America, Asia (except for its western sections) nor all but the north and north-west coasts of Africa. However it is worth remarking that the new worlds being portrayed in the 16th century did not necessarily prompt new charting techniques. Atlantic America and the coastlines of the East were added to the charts by the same portolan chart practitioners as before, initially in the same ways and preserving the same stylistic appearance. Even though huge landmasses were being added to the maps by those who first encountered them, for a considerable period their cartography comprised little more than shorelines. So the medieval chartmakers were merely extending their range – except for the Portuguese, whose nautical cartography was based on astronomical observations)

Top of page

Geometric developments

Uncertainty still surrounds the way in which the coastal outlines were collected, as well as the mathematical projection (if any) used to plot those onto the original chart(s). Later developments are not much clearer. Depictions of the Equator and the tropics, e.g. on the La Cosa chart (1500?) and Cantino planisphere (1502), were followed by the first complete latitude scale (which may occur on one of the undated charts assigned to the first decade of the 16th century). Since that seems to point to the practice of astronomical navigation, even if only implicitly, it can be argued that those are more correctly termed ‘latitude charts’ rather than ‘portolan charts’. 56   As far as traditional portolan charts are concerned Corradino Astengo considered that the addition of latitude derived from Spanish work, and was perhaps first seen ‘in two charts of the Atlantic: that by Conte di Ottomanno Freducci (dated around 1514-15) ... and the 1516 chart by Vesconte Maggiolo'. 57   However, even when a full latitude scale is shown on a portolan chart, it should not be assumed that this was necessarily more than an unconnected addition to outlines that had been replicated in the traditional way. In contrast to the definition used throughout these notes that ‘portolan charts’ are defined by their appearance, Joaquim Alves Gaspar argues (personal communications) that a more precise classification is needed, based on geometry and method of constructions, so as to distinguish the traditional portolan chart from both a ‘latitude’ and a ‘Mercator’ chart.

Two general observations can be made, though even these are contested:

  • the Mediterranean is tilted counter-clockwise at the eastern end by approximately 9 degrees on the portolan charts so that the 36 degree parallel, which should run between Tarifa at the entrance to the Mediterranean and Antioch (Antakya), passes instead through Alexandria. No correction to this consistent tilt has been noted before the second half of the 16th century 58
  • the other general error affecting the Mediterranean was a significant under-estimation of its length. This seems not to have been generally corrected until the 17th century [This comment should be ignored: 8 January 2017.] 59

Toponymic developments

Toponymy is the single most studied aspect of portolan chart history, and for good reason. Almost 3000 continental place names (i.e. not counting the insular ones) have been documented on the charts, from the earliest times up to the 17th century, usually paired with the year of their first appearance on a dated or dateable work. 60  This is an important resource, but one whose limitations need to be understood. The charts’ toponymy was highly dynamic but not necessarily accurate and certainly not, on the whole, up to date. On average, places newly founded or renamed took three generations to reach the charts. 61  However, portolan chart toponymy is a valuable tool in understanding the inter-relationships between the chartmakers and also, perhaps, as a measure of the contemporary geographical knowledge and the priorities of the Mediterranean’s marine communities.

Some specific observations can be made:

Iberian conquests
The occupation by one or other Iberian power of the ports along the coast of North Africa is sometimes acknowledged with their own flag, which can therefore be a useful dating aid. These are what seem to be the more significant ones, with the date range of foreign control:

Asfi (safi) 1488-1551
Casablanca 1515-1755 (nife / anfa had been destroyed in 1468 and ‘Casa Branca’ was built on its ruins after 1515)
Asilah (arzila) 1471-1549
Tangier 1471-1662
Ceuta (seuta) 1415 – present
Peñón de Vélez de la Gomera (bedis) 1508-22
Ghassassa (alcudia) 1506 (briefly)
Melilla 1497 – present

Mers el Kebir (mazalquivir) 1505-1708
Ouahran (oran) 1509-1708
Mostaganem 1506-16
Ténès 1510-12
Algiers [harbour] (zizera, arger) 1510-29
Bejaia (bugia) 1510-51
Annaba (bone) 1535-40

Tabarca Island 1540-1742 [Genoese]
Tunis 1535-69

Tripoli 1510-51 [by the Knights Hospitaller from 1523]

In addition, the Spanish conquest of Naples in 1504 was recognised in a flag showing the Aragonese-Catalan arms. 62

Colonies and trading posts elsewhere
A large number of overseas bases of one kind or another were held in the eastern Adriatic, Peloponnese, Aegean Sea and Black Sea, variously by the Venetians, Genoese or Catalans. An analysis of their toponymy was undertaken to see if they were named on the charts at all and if so whether in red or merely in black. This concluded disappointingly that “the charts displayed little awareness of an overseas possession's growing commercial significance, nor of the point at which its trading importance declined or it became irrelevant once lost to the Ottomans or others.” 63

Newly-named places
A handful of ports, whose foundation or re-naming is recorded, provide a reliable indication that the work in question could not have been produced before that date. 64

Top of page

General observations

  • the portolan charts must have been replicated in many thousands (despite their rare survival – particularly of working examples). Most would have been produced by apprentice-trained chartmakers
  • some of those chartmakers might have done little else all their working lives
  • Pietro Vesconte (fl.1311-c.30) is possibly the first person to pursue mapmaking as a full-time commercial craft. He has been described as ‘the first professional cartographer in western Europe to routinely sign and date his works’ 65
  • what are recognisably portolan charts continue until about 1700, even if their geometric structure was later modified to take note of geodetic advances
  • their large size – up to 1 metre [39 inches] in width – and the need to avoid deterioration of the coastal outlines (at least into the 16th century) required that they were directly copied from a full-size workshop model. Technical descriptions survive from the 16th century but the materials available and methods used in the early period remain conjectural. It is likely that the varied drafting skills involved in the necessarily very precise replication of a portolan chart, sometimes at a different scale, were different to those found in other medieval scribal work and would have required specialised training 66

What do the innovations tell us about the charts’ overall development?

It is tempting to read the introduced elements and the new cartographic techniques described above as evidence of progress. In respect of the improvements that would have aided a navigator, that is probably justified. But the precise and literal procedures for copying the charts, while protecting against degradation of the coastal outlines, simultaneously acted as a general brake on change. Though we refer to their creators as chartmakers, in reality, and with only a few exceptions, they were chart copyists, whose personal contribution is likely to have been restricted, at most, to minor toponymic alterations. It is unlikely that most of them had much understanding of the cartography they were reproducing. [The contribution of the often talented illustrators is another matter entirely.]

In assessing the innovations described above it should always be remembered, that in order to have significance, any new element had to be imitated. If this happened at all – and a few were effectively dead ends and have therefore not been listed – it usually took some time before they were repeated, and even longer before they were generally adopted.

For the first two centuries, during which chartmaking was, broadly, restricted to three centres, there was a certain amount of borrowing between the nearby practitioners in Genoa and Majorca, but rather less between either of those and Venice, a longer sea voyage away. Some of the toponymic innovations of Vesconte, for example, took well over a century to be absorbed into Majorcan charts. 67   Similarly, the updating of the political allegiances conveyed visually via the charts’ flags was carried out so erratically as to lead to the obvious conclusion that it was left to the users to protect themselves against any hostile reception.

The main purposes of this listing therefore are, first, to highlight the practical inventiveness of the early chartmakers (less so those who came later) and, second, to provide an inventory of introduced features that might help with the dating of unsigned works. It is definitely not intended as a developmental narrative of imagined progress from primitive simplicity to some kind of sophisticated precision. Indeed, to some extent the reverse is true. After about 1330, the coastal outlines are more likely to show deterioration because of careless copying than improvements based on new information. And, while clearly dynamic, the charts’ toponymy was not kept systematically up to date, either with regard to fresh developments or the removal of obsolete names. 68

Despite that, as this record of the achievements of the portolan charts demonstrates, they occupy a unique place in the history of cartography.

Top of page

List of the charts consulted above

[For further details on works up to 1500 see A complete chronological listing of portolan charts assigned to the period pre-1501 (an Excel spreadsheet). This cites, inter alia, the Campbell Census numbers and those of the 2007 Pujades DVD]

Carte Pisane: Paris, Bibliothèque nationale de France, Cartes et Plans, B 1118
Cortona chart: Biblioteca dell'Accademia Etrusca di Cortona, Port. 105
Riccardiana chart: Florence, Biblioteca Riccardiana, 3827
Lucca chart: Archivio di Stato, Lucca, Fragmenta Codicum, Sala 40, Cornice 194/1

Pietro Vesconte 1311 chart: Florence, Archivio di Stato, C.N. 1
Pietro Vesconte 1313 atlas: Paris, Bibliothèque nationale de France, Cartes et Plans, DD 687
Perrino Vesconte 1327 chart: Florence, Biblioteca Medicea Laurenziana, Med. Palat. 248
Giovanni Carignano map/chart (pre-1330): Florence, Archivio di Stato, C.N. 2 [destroyed in World War II]
Angelino 'Dalorto' = Dulceti 1330 chart: Florence, Prince Corsini
Angelino Dulceti 1339 chart: Paris, Bibliothèque nationale de France, Cartes et Plans, B 696
Pizzigani brothers 1367 chart: Parma, Archivio di Stato, Biblioteca Palatina, Ms. Parm. 1612
Catalan Atlas (c.1375): Paris, Bibliothèque nationale de France, Manuscrits, MS. Esp. 30
Cresques atelier chart (last quarter of the 14th century – Pujades 2007, DVD: C 19): Naples, Biblioteca Nazionale
      Vittorio Emanuele III, Sala dei MSS 8.2 [ms XII.D 102]
Italian atlas, late 14th or early 15th century: Oxford, Bodleian Library, MS Douce 390
Francesco Beccari 1403 chart: New Haven, Yale University, Beinecke Library, Art Object 1980.158
Sentuzo Pongeto 1404 chart [location unknown, see the 2007 Pujades DVD C 26 for an illustration]
Corbitis and Pinelli-Walckenaer atlases (early 15th century?): Venice, Biblioteca Nazionale Marciana, It. VI,213 (5982);
      and London, British Library, Add. MS 19510
Andrea Bianco 1448 chart: Milan, Biblioteca Ambrosiana, F.260 (Inf. (1)
Petrus Roselli chart 1456: Chicago, Newberry Library, Ayer Collection, Ms map 3
Catalan Estense world map (by Petrus Roselli, c.1462-4): Modena, Biblioteca Estense e Universitaria, C.G.A.1
Petrus Roselli chart 1464: Nuremberg, Germanisches Nationalmuseum, La 4017
Grazioso Benincasa, atlases of 1468 & 1469 [see Tables accompanying the online note: 'The style and content of
      Grazioso Benincasa's charts: Imitation, innovation and repetition']
Jorge de Aguiar 1492 chart: New Haven, Yale University, Beinecke Library, 30cea/1492
Juan de la Cosa chart (1500?): Madrid, Museo Naval, Inv.257
Cantino planisphere (1502): Modena, Biblioteca Estense e Universitaria, C.G.A.2
Conte de Ottomanno Freducci chart (1514-15?): Florence, Archivio di Stato, C.N.15
Vesconte Maggiolo 1516 chart: San Marino, Huntington Library, H.M. 427
Vesconte Maggiolo 1535 chart: Turin, Archivio di Stato, J.b.III.18
Battista Testarossa 1557 atlas: London, Royal Geographical Society, MG265.C.16
Antonio Millo 1586 atlas: Berlin, Staatsbibliothek Preussischer Kulturbesitz, MS Hamilton 446
Thomas Lupo c.1588 chart: London, British Library, Add. MS 10041
Thomas Hood 1596 chart: London, National Maritime Museum, Greenwich, G224:1./2
Gabriel Tatton c.1600 chart: Chicago, Newberry Library, Ayer MS, map 22
Joan Oliva 1616 atlas: Florence, Istituto e Museo di Storia della Scienza, Antico G.f. 25

Top of page


I wish to thank particularly Corradino Astengo, Joaquim Alves Gaspar, Anton Gordyeyev, David Jacoby, Wolfgang Köberer, Jacques Mille, Roel Nicolai, Richard Pflederer, Luis A. Robles Macias, Sarah Tyacke and Chet Van Duzer for their many helpful corrections and suggestions. That does not mean that the commentators agreed with all the assertions above (indeed a few are contested), for which I must take full responsibility. Technical help was received from John Woram.


A number of the links below are to other parts of the ‘Map History’ site’s Portolan Chart pages. For full bibliographical details of any cited reference works see the general ‘Bibliography’

To get from the note number to the endnote text, just click on the number;
to return, click the number or use the back arrow

1. Much of the information in this Note derives from an ongoing investigation into portolan chart history that was started in 2008, when I picked up once more on the earlier work that had resulted in the chapter in Volume 1 of The History of Cartography (Chicago University Press, 1987) [available online Here]. In some instances I cite the update notes to that chapter, since those provide later information and references. The results of the recent detailed analyses of various aspects of the portolan charts, especially their toponymy, continue to be self-published (and updated) on my ‘Map History’ website under the general heading A critical re-examination of portolan charts with a reassessment of their replication and seaboard function (comprising about 30 separate web publications and over 120 tables and graphs). For a summary of the findings set out in those pages see the General Conclusions.

2. First International Workshop on the origin and evolution of portolan charts, whose Abstracts are available online. The main voice disputing a medieval origin was that of Roel Nicolai. [See note 16 for details of his recently published doctoral thesis.]

3. Patrick Gautier Dalché, Carte marine et portulan au XIIe siècle: le "Liber de existencia riveriarum et forma maris nostri Mediterranei" (Pise, circa 1200) (Rome: École française de Rome: distributor, Paris: Boccard, 1995). Collection de l'École Française de Rome 203, p.20.

4. Ramon J. Pujades i Bataller, 'The Pisana Chart: really a primitive portolan chart made in the 13th century?' Cartes et géomatique, 216 (June 2013): 17-32 (paper delivered at the international conference, 'D'une technique à une culture: les cartes marines du XIIIe au XVIIIe siècle', Paris 3 December 2012). The related works are the Cortona and Lucca charts.

5. A detailed reassessment of the Carte Pisane: a late and inferior copy, or the lone survivor from the portolan charts' formative period?

6. This was announced by Catherine Hofmann at the Lisbon workshop (June 2016 – see Note 2) but that finding has not yet been published. For Ramon Pujades's further thoughts on the dating of the Carte Pisane to c.1280-1313, see Els mapamundis baixmedievals: del naixement del mapamundi híbrid a l'ocàs del mapamundi portolà / Late medieval world maps: from the birth of the hybrid to the demise of the portolan mappamundi,(bilingual, Catalan and English), Barcelona: Institut Cartogràfic i Geològic de Catalunya (ICGC), 2023 [but released online in March 2024], pp. 99-101 (121-123 in the online version). {This sentence added 2 June 2024}.

7. Table 2 ’Toponymic time-lag (from physical creation to recognition by mariners)’.

8. Jacques Mille, personal communication 20 August 2016.

9. Vera Armignacco, 'Una carta nautica della Biblioteca dell'Accademia Etrusca di Cortona', Rivista Geografica Italiana 64 (1957): 185-223.

10. Jacques Mille, The French Mediterranean coasts on portolan charts (self-published at the time of the Lisbon Workshop (6-7 June 2016), pp.25-8) [see note 2].

11. Gautier Dalché 1995; Alessandra D. Debanne, Lo Compasso de navegare. Edizione del codice Hamilton 396 con commento linguistico e glossario (Brussels, etc.: Peter Lang for the Gruppo degli italianisti delle Università francofone del Belgio, 2011).

12. Armando Cortesão, The History of Portuguese Cartography, vol. 1, 1969, pp.215-16.

13. Alberto Maghaghi, ‘Nautiche, carte’, in Enciclopedia italiana di scienze, lettere ed arti, originally 36 vols ([Rome]: Istituto Giovanni Trecanni, 1929-39), 24:323-31, especially p.330b.

14. Tony Campbell, A critical re-examination of portolan charts, with a reassessment of their replication and seaboard function (2012-ongoing).

15. Ramon J. Pujades i Bataller, Les cartes portolanes: la representació medieval d'una mar solcada. [In Catalan and Spanish, with English text, 'Portolan charts: the medieval representation of a ploughed sea', pp. 401-526.] (Barcelona: Institut Cartogràfic de Catalunya; Institut d'Estudis Catalans; Institut Europeu de la Mediterrània; Lunwerg, 2007).

16. Roel Nicolai, The Enigma of the Origin of Portolan Charts: A Geodetic Analysis of the Hypothesis of a Medieval Origin (Leiden: Brill, 2016). This sentence corrected in the light of Nicolai's comment (in a post to the ISHMap-List, 4 January 2017): 'My position is not that they “have been passed down in their completed form”, that is, as complete charts of the Mediterranean, but that they were passed down as partial charts from an earlier period, which indeed has not been identified yet. I have proposed that these partial charts were copied and pasted together in medieval times to create a mosaic chart of the Mediterranean, the Black Sea and the Atlantic coasts'. The Atlantic coasts were of course missing from the earliest surviving charts.

16a. See ‘Mediterranean portolan charts: their origin in the mental maps of medieval sailors, their function and their early development’ {published online on 26 January 2021}.

17. See, for examples, P.D.A.Harvey, Medieval Maps (London: British Library, 1991); and also the comment about the al-Idrisi map (Note 20).

18. The mathematical device designed to calculate both the effective distance gained at sea when the ship had been forced off course, and the new optimum direction involved. See the relevant update notes to the 1987 chapter, ‘Toleta di marteloio’.

19. A deeper understanding of the function of portolan charts.

20. Although the al-Idrisi world map of 1154 (on which see Konrad Miller’s redrawing on Wikipedia), and indeed the Ptolemaic maps, include some landforms that are broadly recognizable, if far less realistic.

21. Again, the al-Idrisi map does include several large gulfs set into the North African coast, but not recognisable outlines of the two discussed.

21a. The names of ports and anchorages are written inland, and sometimes at right angles to the coast, on Islamic charts, for example those of the 10th-century geographer Ibn Hawqal. [I owe this observation to Stefan Schröder]. {This endnote added 17 November 2023.}

22. Red Names on the Portolan Charts (1311-1677): a detailed investigation; and, specifically for those three statements: Human and physical features, and the next section ‘Relationship of black and red names’; and, for the chartmakers involved: Summary Table of Red Names: their appearance, frequency and disappearance (sort on column 9 ‘Chartmaker first to show in red’).

23. I owe this observation to Leif Isaksen. One exception to that generalization is the 1318 Vesconte atlas (the one in Vienna), which places an overall yellow tint over the sea areas. However, this is sufficiently muted so that no written details are obscured. Another exception, and a reflection of its non-nautical purpose, is the densely patterned sea area on the Catalan Atlas (c.1375).

24. Pujades 2007, p.220.

25. Table 7 ‘Development of the signs for navigational dangers’.

26. Richard L. Pflederer, Finding their way at sea: the story of portolan charts, the cartographers who drew them and the mariners who sailed by them ('t Goy-Houten: HES & De Graaf, 2012), p.17.

27. Campbell 1987 [see Note 1], p.378. We can perhaps take comfort from those who study lichens, one of whom pointed out they were involved in "an entire discipline that can't define what it is they study" (Quoted in Merlin Sheldrake, Entangled Life (2020), 101.

28. Corradino Astengo, 'The Renaissance chart tradition in the Mediterranean', in: David Woodward (ed.) The History of Cartography. Volume 3. Cartography in the European Renaissance (Chicago & London: University of Chicago Press, 2007), Part 1, pp.199-203, including a brief overview of city views on the charts over the centuries.

29. The meaning of the different elements referred to would have had to be explained, first by the chartmaker to his mariner purchaser, and thence to the sailor’s colleagues and apprentices. Likewise, whatever route the toponyms had initially taken on their way onto the chart, any subsequent addition or correction must have passed in the opposite direction – orally from the mariner to the chartmaker – since none of the practitioners before the 15th century (as far as we know) had sailing experience themselves. Francesco Beccari refers in his famous 'Statement to the Reader' (transcribed but not translated in Pujades (2007, p.461)) how he had obtained his information from masters, ship-owners, sailors proficient in navigation, and pilots. Prior to the portolan charts, a seaman’s experience must have been built up from memorised oral instruction. The place of orality in the creation, updating and use of the portolan charts has not hitherto been sufficiently recognised .

30. For a full listing of surviving portolan charts and atlases, see Richard L. Pflederer, Census of Portolan Charts & Atlases (Privately published, 2016 – an amended version of the original 2009 edition, now with over 2000 entries, available in digital form (Excel) from the author: richard (at) pflederer.net). For illustrations of a wide range of charts through the centuries from the unparalleled collection of the Bibliothèque nationale de France, see Monique de La Roncière & Michel Mollat du Jourdin. Les Portulans: Cartes marines du XIIIe au XVIIe siècle (Fribourg: Office du Livre, 1984). [English edition, translated by L. le R. Dethan, Sea charts of the early explorers: 13th to 17th century (London: Thames & Hudson, 1984)]; and also Vicenç M. Rosselló i Verger, Portolans procedents de col.leccions espanyoles: Segles XV-XVII [catalogue of an exhibition at the Salò Tinell, Barcelona, on the occasion of the 17th conference of the International Cartographic Association, 5-17 September 1995] (Barcelona: Institut Cartogràfic de Catalunya, 1995).

31. Colours used for the islands, estuaries, etc. However, Jacques Mille has pointed out that the Avignon chart, which may well pre-date Vesconte, includes coloured islands {Note added July 2020}.

32. Tony Campbell, ‘Why the artificial shapes for the smaller islands on the portolan charts (1330-1600) help to clarify their navigational use’, Cartes et géomatique, 216 (June 2013): 47-65 (paper delivered at the international conference, 'D'une technique à une culture: les cartes marines du XIIIe au XVIIIe siècle', Paris 3 December 2012); and Island shapes as a mnemonic device .

33. Campbell 1987 [see Note 1], pp.398-401, and the update note to that chapter [scroll down to ‘Flags’], which includes references to various flags that can be used for dating a chart. [A detailed study of the portolan charts’ vexillology is long overdue.] However, it needs to be remembered that while the chartmakers might speedily acknowledge Christian victories, such as the conquest of Granada in 1492, Ottoman advances were commonly ignored: the notable example being Constantinople’s fall in 1453. On flags, see also ‘Iberian conquests’.

34. For details see Listing and analysis of portolan chart toponyms along the continuous coastline from Dunkirk to Mogador (early 14th to late 17th century) including the transcribed names from the 'Liber de existencia riveriarum' and 'Lo compasso de navegare' as well as the Carte Pisane and Cortona chart (sort on column G, then F).

35. For illustrations see Pujades 2007, p.220.

36. For illustrations of these various features see Pujades 2007, pp.228-35.

37. Heinrich Winter, 'A Late Portolan Chart at Madrid and Late Portolan Charts in General', Imago Mundi: the International Journal for the History of Cartography 7 (1950): 37-46, see pp.37-40. See also Campbell 1987 [see Note 1], p.395; and the update note to that chapter (‘Compass rose’). Pujades 2007, p.235 illustrates a range of styles.

38. Name labels. That this device is found on the chart sold at Sotheby’s on 17 June 1997 is just one of the features contradicting its earlier attribution to Vesconte, see Census additional entries, E.16.

39. Philipp Billion, 'A newly discovered fragment from the Lucca Archives, Italy', Imago Mundi 63: 1 (2011): 1-21 & coloured plates 1-2, see pp.9-11.

40. René Tebel, Das Schiff im Kartenbild des Mittelalters und der Frühen Neuzeit: kartographische Zeugnisse aus sieben Jahrhunderten als maritimhistorische Bildquellen. Schriften des Deutschen Schiffahrtsmuseums 66 (Bremerhaven: Deutsches Schiffahrtsmuseum; Wiefelstede: Oceanum, 2012), p.120.

41. Pujades 2007, p.461 where the 'Statement to the Reader' is transcribed (but not translated from Latin).

42. Campbell 1987 [see Note 1], p.396.

43. This was noted first on Gabriel Tatton’s traditional portolan chart of c.1600, and became a regular feature of the Thameside Drapers’ School.

44. Portolan atlases would often have one chart stuck to the back of another (see Astengo 2007, pp.182-5 for descriptions of different formats). What may have been a late 16th-century development, much favoured by the English ‘Thames’ School, was to lay a single chart down onto hinged oak boards (in two or four sections), so that it could be folded away for protection on board ship. Many of the charts involved have been removed from their boards but the cut-outs for the hinges confirm their original format. Hinged boards were used on a chart dateable to c.1588 signed by Thomas Lupo in London and now in the British Library (information from Sarah Tyacke). However, the use of the boards for the arms of a Medici cardinal in that instance shows that this was presumably a presentational device, not for shipboard storage.
     A chart of 1596 by Thomas Hood (National Maritime Museum), which may be the earliest surviving dated example of boards designed for use at sea, was followed by others from Gabriel Tatton and John Daniel. I owe these facts to Richard Pflederer (personal communication). It might still have originally been an Italian device. Astengo 2007, p.184 illustrates a 1535 chart in four sections by Vesconte Maggiolo (Archivo di Stato, Turin): ‘however, we have no way of knowing if this was the way they were originally mounted’.
     Eva Oledzka (see here) has described the wooden boards and outer leather container of the Bodleian Library's Italian atlas of around 1400 (MS Douce 390). That this also appeared to have had a carrying strap suggests that even such a prestige item might have had practical use. {This paragraph added 30 November 2016}.
     Pflederer also drew my attention to a group of charts by different authors in the Biblioteca Nazionale Centrale, Florence, which – reversing the usual sequence – had been uniformly mounted in that way much later, perhaps so that they could be shelved like books.

45. Campbell 1987 [see Note 1], pp.409-10; and Heinrich Winter, ‘The changing face of Scandinavia and the Baltic in cartography up to 1532’, Imago Mundi 12 (1955) pp.45-54 (which concentrates mostly on the 16th century). See also a note on the Catalan Estense world map.

46. E.3. North Sea coasts (Denmark to Dieppe); and Carte Pisane Hydrography Tables – Table 3 ‘Early North Sea names’. Table 4 ‘Name totals between Bruges and Seville’, and Table 5 ‘Red names from Calais to Seville’, analyse the rest of the Atlantic coastline.

47. Carte Pisane Hydrography Tables – Table 1 ‘Development in the outline and toponymy of the British Isles’. See also Table 2 ‘Early names along the South coast of England’.

48. The features chosen to illustrate the developing portrayal of islands and estuaries on portolan charts up to 1469 (nos 1 & 2).

49. The features chosen to illustrate the developing portrayal of islands and estuaries on portolan charts up to 1469 (nos 9 & 10).

50. Armando Cortesão, History of Portuguese Cartography, 2 vols (Coimbra, 1969-71), 2:184-8; see also The style and content of Grazioso Benincasa's charts: imitation, innovation and repetition – Table 2, penultimate column, ‘Cape Verde Islands’.

51. On Brasil see Barbara Freitag, Hy Brasil: the metamorphosis of an island from cartographic error to Celtic Elysium. Textxet. Studies in Comparative Literature 69 (Amsterdam & New York: Rodopi, 2013); and, on Antillia, the Wikipedia entry (checked 11 September 2016), which needs more up-to-date references.

52. See the table in Campbell 1987 [see Note 1], pp.412-3; and the update note to that chapter [scroll down to ‘Western coast of Africa’]. For the argument that Petrus Roselli seems to have had access, by 1464, to the toponyms deriving from Portuguese voyages up to 1461, which Benincasa was able to include in full only in 1468 see The style and content of Grazioso Benincasa's charts: imitation, innovation and repetition (West Africa).

53. Gautier Dalché 1995, p.40; Debanne 2011, p.120 – referring to folio101r of the original.

54. E.6. The Black Sea.

55. Cyrus Alai, General Maps of Persia 1477-1925 (Leiden: Brill, 2005), p.3.

56. Joaquim Alves Gaspar, 'From the portolan chart to the latitude chart: the silent cartographic revolution', Cartes et géomatique, 216 (June 2013): 67-77, especially p.70, note 12 (paper delivered at the international conference, 'D'une technique à une culture: les cartes marines du XIIIe au XVIIIe siècle', Paris 3 December 2012); see also Campbell 1987 [see Note 1], p.386 and the update note to that chapter . Among undated charts from the first decade of the 16th century are the Caverio, King-Hamy and Kunstmann III, as well as that by Pedro Reinel of c.1504 (personal communication from Joaquim Alves Gaspar, August 2016).

57. Astengo 2007, pp. 174-262, see p.193b. While most commentators (including myself) have considered the latitude scale on the Francesco Beccari chart of 1403 to be a later addition, Fortunato Lepore, Marco Piccardi & Enzo Pranzini, Costa e arcipelago toscano nel Kitab i Bahriye di Piri Reis. Un confronto cartografico (secoli XIII-XVII) (Pisa: Felici Editore, 2011), pp.129-35, argue that it is an original feature. For my counter argument see ’Latitude scale’.

58. Portuguese astronomically-based charts from about 1525 were the first to correct the tilt, see Joaquim Alves Gaspar, ‘Dead reckoning and magnetic declination: unveiling the mystery of portolan charts’, e-Perimetron, 3:4 (2008), pp.191-203, especially pp.200-01. For Mediterranean-produced charts, the first seems to be Battista Testarossa in 1557, followed by Antonio Millo in 1586. A more general recognition had to wait for the beginning of the next century. Joan Oliva in his atlas of 1616 shows the East Mediterranean in two sheets, one rotated according to tradition, the other corrected (personal communication from Corradino Astengo, August 2016).

59. [It has been pointed out, by Luis A. Robles Macias and Roel Nicolai, that this comment is incorrect and should be removed. The issue is a complex one. 8 January 2017]

60. Listing and analysis of portolan chart toponyms along the continuous coastline from Dunkirk to Mogador (sort on column F then G).

61. Table B. ‘Toponymic time-lag (from physical creation to recognition by mariners)’.

62. The Pedro Reinel chart in Bordeaux, Archives Départementales de la Gironde, II.Z. 1582 bis; see the note about this on the ‘Census’ update page (see No.5).

63. Red names of overseas trading-posts; and for the accompanying table, List of colonies and trading posts.

64. Carte Pisane Specific Names Tables ‘B. Historical time-lag (from physical creation to recognition by mariners)’. This deals with the earlier centuries only; it needs to be extended to the 16th and 17th.

65. David Woodward, ‘Medieval Mappaemundi’, in: J.B. Harley & David Woodward (eds), The History of Cartography. Volume 1 (University of Chicago Press, 1987), pp. 286-370, especially p.314 [available online Here].

66. Stages in the construction of a chart. For a present-day reconstruction of a portolan chart see Kevin Sheehan, 'The Functions of Portolan Maps: an evaluation of the utility of manuscript nautical cartography from the thirteenth through sixteenth centuries', 360-page PhD dissertation, Durham University, July 2014.

67. Adriatic reappearances (and Catalonia & Valencia).

68. Innovative Portolan Chart Names, see, e.g. ‘Historical time-lag’ and ‘Dissemination time-lag’.

Top of page

To get from the note number to the endnote text, and vice versa, just double-click on the number
For the current locations of the charts referred to in this Note see List of the charts consulted

HOME (main menu)  |  Tony Campbell articles  |  Portolan charts menu  |  Index   |   Sitemap   |   What this site is about   |   What's New