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

The features chosen to illustrate the developing portrayal of islands and estuaries
on portolan charts up to 1469

Mounted on the web 7 March 2011

Visual indexes to the Colour & Shape Analysis   |  Summary table of the features analysed (Microsoft Word doc.)

Colour & Shape Analysis Menu   |  Portolan Charts Main Menu

This prints out to about 13 pages

Conventions and abbreviations used in the Colour & Shape Analysis tables

Conventions and abbreviations used in the tables

    Text abbreviations:

    |     area not covered
    | |     different tints on different sheets
    X     omitted (feature or colour)
    ——     scan not available
    btn     button (with the number of holes)
    (Ci)     circle
    col(s)     coloured/colours
    (Cr)     cross
    (D)     diamond
    (E)     edging
    (H)       hourglass
    i'd       island
    is       islands
    + L       with uncoloured strip left for a name label
    (R)       rectangle
    (SP)       spanner head
    uncol       uncoloured
    (W)      wash (overall)

    Colour abbreviations:

    Bk    Black
    Be    Blue
    Br     Brown
    Go    Gold
    Gn    Green
    Gy    Grey
    M     Mauve [including Purple]
    O    Orange
    Pi    Pink
    R    Red
    Si    Silver
    W    White
    Y    Yellow

The British Isles and surrounding islands
(to identify obscure features see the Visual index)

1. till, a large, mythical circular island off the north-east coast of Scotland [apparently identified with Ultima Thule].
Introduced by Dalorto/Dulceti in 1330 (in green), it was repeated on later Catalan charts in blue [occasionally in green] (from 1375 until at least 1468). Dalorto included within the circle the outline of a church (right-reading on the basis of north orientation). This is apparently not present on the 1339 and undated Dulceti charts. The building can be seen (if faintly) on the 1367 Pizzigani chart, but this time 'upside down', within an uncoloured circle. When the circle was repeated by Francesco Beccari in 1403 (without the church) it showed the red wash invariably used by Benincasa. Subsequent Italian charts employed a variety of colours, with blue found only on the 1455 Pareto chart. {This section amended 6 October 2011. I am grateful for the prompting from Ramon Pujades}

       Button: The Catalan Atlas introduced the concept of showing the circular island as a button. Virga (1409) presented that as a multi-coloured button, with five gold 'holes' surrounded by coloured dots. Macià Viladesters included a simple blue circle with the five holes in 1413, then dropped the holes for 1423, while Joan Viladesters turned till (and montorio/brazil) into a spoked wheel in 1428.

      Label: Benincasa consistently left an uncoloured band as a label for the name isula de till, a form not observed on any earlier chart but one that was imitated later.

2. scurçe, an island off the south-west coast of Scotland shaped like an American football, also mythical.
Again introduced by Dalorto/Dulceti in 1330 (in red), it was repeated, usually in that colour, by subsequent Catalan chartmakers. Surprisingly, there was only one (possible) previous sighting of Benincasa's consistently used green and that on the 1367 Pizzigani chart, which marks the first Italian appearance of the island. Since that doubtful instance was a century before Benincasa, it is unlikely to have served as his model. On this occasion, Benincasa certainly did not copy from his Beccari forbears, who used red, blue and gold respectively on their three charts.

     Dual Colour: Ziroldi (in 1422 and subsequently) and his fellow Venetians, Briaticho and Nicolai divided the island vertically into two, differently coloured, sections. That feature has been noted exclusively on Venetian work in the period 1422-48.

     Label: A similar feature to that on till was added by Benincasa in 1461. This feature has been noted only on the work of Benincasa, his successors and copyists, up to at least 1555.

3. Isle of Man
One of the larger British islands, midway between Ireland and England, this runs south-west/north-east and is broadly oval in shape. Most chartmakers, however, treated it geometrically, as a cross with four even arms (roughly the Occitan form but without the three small circles at the end of each arm - as used also for inferno/Tenerife). This is unrelated to the island's arms, the triskelion, formed of three bent legs with a spur, which reflected ownership by the Norwegian kings until 1266.

The cross form first appeared on three of the Vesconte works dating from or about 1321, but was not adopted by Catalan chartmakers for over a century. Prior to that, the Catalans used two alternative outlines. The first, found only on the 1330 and 1339 Dalorto/Dulceti charts, attempted realism with a (broadly) rectangular shape. However, a version of that was revived in 1424 by Pizzigano and is then seen also on the Medici Atlas and another undated Venetian work (Pujades A 26), both of which are assigned to the period 1425-50.

The second Catalan alternative, introduced onto the late, undated Dulceti chart (post-1339), is roughly rectangular but pinched in the middle so that it looks a bit like an hourglass. That shape reappears on Catalan charts (only) up to 1428. From the 1439 Vallseca onwards all the infrequent Catalan depictions of the British Isles abandon the semi-realistic outline for the Isle of Man in favour of the purely symbolic cross. This is the only noted instance of such a reversion. From then onwards, all other charts up to 1469 (wherever produced) display the cross, with two unique exceptions, each on an unsigned work: a complex jigsaw outline on Pujades A 9 (Genoese, c.1325-50) and an arrangement of four dots on his C 50 (Italian, c.1425-50). See also see under Italian: second quarter 15th century in 'Anonymous works and the question of their attribution to individual chartmakers or to their supposed workshops'.

4. Isles of Scilly
It was not surprising that this strategic but complex archipelago guarding the western mouth of the English Channel should have caused difficulties for the early chartmakers. They were still a problem to 18th-century British hydrographers.

In reality the group comprises one large island, with other medium and small islands to the north, west and south of it. Vesconte was the first to attempt a representation, in 1318. He showed, all in red, a banana-shaped island to the east and a semi-circle of small islands to the west of it. Dulceti (1339) was still repeating this interpretation, again in red. The 1373 Pizzigani treatment is unclear but by Soler's chart of 1385 the main island shape had become more like the head of a traditional spanner.

The spanner form was picked up by F. Beccari (1403) and appears on Catalan (and some Italian) charts thereafter. Some of the Italian charts show variations. Virga (1409) reverses the spanner shape, while Pizzigano (1424) and Bianco show two large islands rather than the original one.

From 1403 onwards the usual convention was to colour the two elements differently. One or other (sometimes both) would be in red but there was no consistency about the second colour. Strangely, given the variety on offer, no previous instance was noted of Benincasa's invariable red and green combination. This can therefore be considered as part of his unique visual signature. Benincasa's final surviving work, the magnificent 1482 chart, showed no improvement over Soler's spanner outline for the main island, introduced a century earlier.

A convention noticed first on the Aguiar chart of 1492 was to abandon the attempt to portray the real archipelago, replacing the outlines with a series of dots. Several further examples of this were seen up to at least the mid-16th century.

5. Isle of Wight
Given its erratic use, colour is of little diagnostic value here until Roselli and Benincasa both consistently repeat the early Catalan preference for red.

6-8. Ireland, Scotland and England & Wales
In the 14th century there are just a couple of instances, both Catalan, of colour wash being used to distinguish the component parts of the British Isles - as would also be done sometimes for Corsica, Sardinia and Sicily. The 1339 Dulceti chart covered Ireland and Scotland with the same light overall wash, leaving England untouched, whereas the Catalan Atlas used a brown wash for Ireland but covered Britain as a whole in mauve. There are no equivalent 14th-century examples and the only later instance is on an anonymous Venetian chart (Pujades C 28), attributed by him to Virga at the beginning of the 15th century, which gives each of the component part of the British Isles the same brown wash.

A band of colour to edge the coastline of one or more of the entities was first seen in the 1420s, initially in 1426 on the Batista Beccari chart (just England & Wales) and one of the same date by Ziroldi (just Ireland and Scotland), while the 1428 Viladesters chart edges just England & Wales in green. Roselli (from 1456) used a consistent pink or mauve for Ireland and green for England & Wales (leaving Scotland untouched) but it was Benincasa who first edged all three entities in different colours: respectively, green, brown and pink/mauve.

See also The presence of an armorial device within the separate countries of the British Isles, as well as within Corsica, Sardinia and Sicily.

9. montorio/brazil, a large, mythical circular island off the south-west of Ireland. Note that the name for this and following could vary or be confused.

     Colour: this was variously and inconsistently coloured after its introduction on the 1330 Dalorto/Dulceti chart, apart from Roselli's predictable mauve/pink/red.

     Button: this feature appears on the ornate Catalan Atlas (9-hole button), variants of which are seen on other Cresques atelier productions, and the 1428 Viladesters chart (spoked wheel).

      Label: an uncoloured band as a label for the name was left by the Pizzigani (1367 & 1373) and by Cesanis (1421, who also makes the edge serrated), before Benincasa made this combination an invariable feature of his work. One or two unsigned 15th-century Venetian productions include the label and serration, after which it is the exclusive hallmark of the Benincasas, their copyists and successors up to at least 1563.

10. man, another, fairly large, mythical island, this time half-moon-shaped, to the south of montorio. [Not to be confused with the Isle of Man in the Irish Sea - No.3.]

Introduced on the 1367 Pizzigani chart, and included thereafter when the coverage allowed, this was variously coloured until Roselli and Benincasa both chose to use a consistent blue.

An uncoloured band around an inner blue strip first appeared on one of the undated Cresques atelier works assigned to the last quarter of the 14th century (C 19). It was repeated on two Viladesters charts (1423 and 1428). Two undated Venetian atlases assigned to the second quarter of the 15th century (A 30 & A 31) introduce an uncoloured label strip, a convention followed consistently by Benincasa. This feature forms part of the unique signature of Benincasa, his copyists and successors (up to at least 1563), since it has not been seen on the work of other chartmakers.

Top of page

Canary Islands
(for identifications see the Visual index)

11-13. The two main easterly islands, Lanzarote and Fuerteventura, appear on the 1339 Dulceti chart. The entire archipelago is first shown on the 1367 Pizzigano chart. There are no clear pattterns in the colour choices, except for Benincasa. Most interest lies in the treatment of Tenerife (inferno) and Lanzerote. No comments are made below about 11 Hierro [ferro], 12 [La] Palma, 13 Gomera or 15 [Gran] Canaria.

14. Tenerife (inferno)
A white device was sometimes set into the wash tint for the island (often red), clearly intended to represent the snow-covered Teide-Pico Viejo volcano complex, which dominates the island's centre. Shortly after the island's first appearance on the 1367 Pizzigano chart the Catalan Atlas added an indented circle (like a cog-wheel), which seems to have been repeated on the three Beccari charts (1403, 1426 & 1435) and that by 1423 Macià Viladesters. Occasional works, including Briaticho (1430), used a white circle. His fellow-Venetian Bianco (1436 & 1448) added an inner red circle. The 1413 Macià Viladesters chart has a (difficult to read) star-shaped pattern.

The final variant is first seen on a chart attributed to Rafel Soler in the second quarter of the 15th century. His single signed work had not included the Canaries. It is unclear if Soler's symbol is meant to represent a four-leaf clover or a cross, and the confusion deepens when Benincasa adopted the device in 1461. On his earliest works it looks more like a clover, which is endorsed by the five-leaf form on the Florence chart (C.N.6) - assumed to have been produced in the same year as his first chart, 1461. Clearly, that could not be a cross. Thereafter, where most instances look more like a cross, a few could be clover. If what is placed in the middle of the red island is a cross it is a version with four even arms, roughly the Occitan form but without the three small circles at the end of each arm - thus similar to the outline used for the Isle of Man (No.3) but distinct from that denoting Paxos (No.31). Few post-1469 Benincasa charts are available in reproduction but it was possible to see in the 1476 Geneva atlas signed by Andrea Benincasa a device reminiscent of his father's early five-leaf form. In this case there are five small circles each with a central dot, like frog-spawn.

The distinguishing device could be mistakenly applied to the neighbouring Gran Canaria [15], as seems to be the case on the 1508 Russo chart, while the clover-shape is clearly misplaced on the 1563 Sideri atlas.

16 fuerteventura
The Pinelli-Walckenaer Atlas (late 14th or early 15th century) is alone in placing a white circle in the middle of the island. This was presumably a mistake for Tenerife.

17. lanzerote
From the time of its first appearance in 1339 this was marked out with a cross, usually in red on an uncoloured (or, less frequently, black) background. Occasionally it was the cross that was white, for example on the 1430 Briaticho and 1436 Bianco atlases. There were few exceptions to this, just the Venetians: Cesanis, Ziroldi and Nicolai (whose relevant works covered the period 1421-48). They omitted the cross, as does the chart attributed to R. Soler (Pujades C 51). Harvey (2009, p.51) refers to the 'red cross of Genoa - perhaps the only unquestionably symbolic colouring to appear on medieval maps'.

Top of page

Balearic Islands
(for identifications see the Visual index)

Shown in colour from the time of Vesconte's first atlas (1313), the choice of tint showed an unusual level of agreement among chartmakers, for example:

18. Formentera - consistently red on the work of Vesconte, Dulceti, the late-14th-century Majorcans, the Beccaris, Viladesters, Vallseca, Roselli and Benincasa. The exceptions were the Venetians, most of whom favoured blue, green or brown

19. Ibiza - after Vesconte's green, this was mostly treated in blue, the exceptions again being most of the Venetian practitioners

20. Majorca
      Shape: see Colour & Shape Notes (Majorca)

      Colour (island): Vesconte started by applying an overall wash to the island but, by 1321 that had been dropped. It is possible to see some brown wash on the 1330 Dalorto/Dulceti chart but that may be an illusion. The 1367 Pizzigani chart then re-introduced the convention of colouring the island that would be followed on almost all later charts.

      Colour (stripes): The Catalan Atlas was the first to use the island to display the Aragonese colours, with its alternating red and gold stripes. However, that effect is repeated on only one of the four charts attributed to a Cresques atelier, as well as the 1413 Viladesters chart, and it does not become usual Catalan practice until the work of Vallseca (possibly with the gold sometimes replaced with silver - the chromatic reading is not always clear).There was just one Italian instance (1426 Beccari) before it became a standard feature of Benincasa's work.

      Stripes (number and orientation): In relation to north these are found roughly running in line with the eight main compass directions, e.g. \ | / —. The line of the island's broadly parallel east and west coasts invited stripes that ran either south-west/north-east or at right-angles to that. Following the example of the Catalan Atlas most Majorcan work repeated the south-east/north-west stripes. The number of red stripes varied on these charts from four to eight. No significance seems to have attached to the number - Roselli used five, six or seven at different times - but that orientation is significant as a pointer to Catalan work, since it was never (apparently) used by Italian chartmakers.

Rafel Soler introduced, and Benincasa would consisently use, two stripes in the opposite direction. Finally, there are examples of four vertical stripes (which ignored the island's shape) on a chart assigned to the Cresques atelier (Pujades C 22), while six such stripes are found on the 1426 Beccari chart, as well, surprisingly, on Vallseca's chart of 1449. Roselli's eight (careless) horizontal stripes (1447) are also unexpected, though that is his earliest surviving chart.

      Label: In place of the stripes of Aragon, the two early Genoese productions in Paris (MS Vat.4850 & MS Ital.1704) left a white space on which the island name was written. These strips, which run along the island's long axis, were aligned in different directions on the atlas and chart respectively. Later, the Venetians - Pizzigani (1373), Cesanis, Ziroldi, Nicolai and Fiorino (who included several names), as well as some unsigned works - placed their white strips in a consistent way, linking the two big bays. Pareto employed a white strip that was closer to that of his Genoese forebears. Despite these small differences in the placement of the name label, this provides an almost exclusively Italian feature to match the Catalan's stripes, since only two anonymous works assigned to the Cresques atelier display a name label (Pujades C 15 & C 18).

Finally, three related atlases, assigned to the second quarter of the 15th century by Pujades (A 29, 30 & 31) include multiple labels across Majorca - seven, six and three respectively - thus emphasising the preference for this device among Venetian practitioners.

21. Minorca
The 14th-century Genoese, and the Catalan chartmakers in general, used red with remarkable consistency. The Venetians used various colours and Benincasa an (almost) invariable green, for which only two precedents were noted: as an alternative treatment on an unsigned Italian atlas (beginning 15th century? - Pujades A 13) and, possibly, on the Maghreb chart (C 54).

Top of page

Central Mediterranean
(to identify obscure features see the Visual index)

22 & 24. Islands in the Rhône Delta
The long sandbank, roughly between Sète and Aigues-Mortes, and the two lakes it encloses, was a challenge to portolan chartmakers. Vesconte in 1313 was to the first to grapple with it and showed a thin red island with a whale-shaped area behind (to the north of it) stippled red. The Rhône delta to the east [24] is conveyed by a single triangular island. By 1321 Vesconte was showing two delta islands.

In 1330, Dalorto/Dulceti joined the thin island up to one of the larger delta islands, giving it the shape of a pre-industrial plough, with one or two further islands to the south-east (coloured differently or the same). The variety is too great for meaningful conclusions. For eleven comparative outlines (Carte Pisane to the 1489 Bertran chart), first published by Oldham in 1925, see Gautier Dalché (2001, p.22), accessible online via Google Books.

23. Aigues-Mortes
Another 1330 addition is diagnostically more useful. Dalorto/Dulceti separated the initial red 'a' of Aigues-Mortes and placed it inside that new, plough-shaped island - here uncoloured, where the section to the south is washed green. For an illustration see the Scans page. Though there was no general consistency in this, Catalan chartmakers tended to detach the first letter, as did the Beccaris (though the 1403 chart repeats the initial letter) and (part of the time) Ziroldi. The convention appears infrequently on Venetian work. For both Roselli and Benincasa it was standard. Rafel Soler, uniquely on his signed Berlin chart, placed the first two letters, 'ay ', on the island. For a similar treatment of the 'd' of damiata (see Nile delta).

The detached 'a' convention was last noted on a non-Benincasan style chart in 1505. However, there is a hint of it in the way that the 1520 Xenodocos atlas omits the initial letter, so that the name starts ygues.

25-27. Corsica, Sardinia and Sicily
On his two later charts, Dulceti applied overall washes to distinguish the three islands, as did just one of the five early Genoese productions (the atlas in Paris, MS Lat.4850 - Pujades A 9). The Catalan Atlas covered each island in gold with a swirling red pattern on top (the 1413 Viladesters just added the gold). Two anonymous Italian charts from the first half of the 15th century were the only other non-Catalan works to apply overall wash to all three islands, the second of them using different colours for each (Pujades C 31 & C 50).

In 1426 Ziroldi decided to edge all three islands, though using just two colours. Battista Beccari, certainly by 1435, was the first to distinguish each island outline in a different colour. Since few chartmakers over the next 30 years coloured up all three islands, Benincasa may have taken the idea from Beccari, although, if so, the former's almost invariable colours (blue for Corsica, pink for Sardinia and green for Sicily) were different from those of Beccari's. A few other chartmakers edged one or two of the islands or, if all three, with one colour repeated.

See also The presence of an armorial device within the separate countries of the British Isles, as well as within Corsica, Sardinia and Sicily.

28. Malta
The colouring for Malta was very inconsistent apart from Vesconte (green), Vallseca and Roselli (blue), Ziroldi and Benincasa (gold).

29. Gozo
Gozo, on the other hand, was red in the great majority of identified instances (75, against 17 definitely in another colour). All but two of those 17 exceptions were Venetian productions using blue, found on the work of Ziroldi, Briaticho and Nicolai. The final exceptions were Virga's brown and the 1449 Vallseca chart's gold.

Top of page

Ionian Islands
(for identifications see the Visual index)

Shape and evidence of colour inconsistency in the Ionian and Aegean islands

In general, because of the practical difficulties of reproducing or describing shape differences, that aspect was not analysed. However, an exception was made for certain Ionian and Aegean islands, where marked differences in outlines provide a useful diagnostic tool - see The different approaches to conveying shape: Islands.

Because the Ionian Islands sometimes appear twice, on the central Mediterranean (or Adriatic) and Aegean sheets, these can provide unequivocal evidence of internal inconsistency, when the four larger islands considered here were tinted differently. This can be seen, and is shown in the analysis by the convention ||, on the work of Vesconte (who tinted Zakynthos three different tones on the atlas in Lyons), Pizzigano (1373, with all four pairs differently coloured), on the 1446 Ziroldi chart and an atlas attributed to him, on the 1448 Nicolai atlas, and the anonymous early Genoese atlas in Paris (Pujades A 9).

A change in the colour of one island in a group could well lead to a whole new sequence of colours. In Vesconte's case this varied treatment can be seen in the way he used colour to give a very different appearance to Rhodes on charts 3 and 4 of his 1313 atlas. [Similar inconsistency can be seen in the colours applied to the main Balearic islands on two sheets of the unsigned 'Italian' atlas of c.1325-50 (Pujades A 13).]

30. corfu (Kerkyra)
      Colour: the choice of colour was generally inconsistent, except for Vallseca's and Roselli's invariable decision to leave it uncoloured.

      Shape: see Colour & Shape Notes (Corfu)

31. Paxos
      Colour: in a rare display of conformity Catalan charts invariably coloured it red, as did Vesconte and Benincasa. There was no agreement among other Italian chartmakers.

      Shape: see Colour & Shape Notes (Cross shape)

32. sca maura (Lefkada)
      Colour: 14th-century Catalan charts coloured it red, Ziroldi used blue, and Roselli and Benincasa green.

33. ithaca (Ithaki)
This was omitted from the Catalan Atlas, the 1456 Roselli chart and two Ziroldi atlases (1456) as well as an atlas attributed to him (Pujades A 27). Is this perhaps evidence that Ithaca, even if not the other Ionian Islands, did not form part of the workshop model used for tracing or pouncing and would instead be drawn in freehand?

      Colour: Vesconte and Benincasa used red, as did Roselli and most 15th-century Catalan charts.

34. cefalonia (Kefallonia)
      Colour: this was blue or uncoloured on virtually all Catalan charts.

35. jacinto (Zakynthos)
      Colour: almost all Catalan charts used red.

      Shape: see Colour & Shape Notes (Zakynthos)

Top of page

(to identify obscure features see the Visual index)

36. stalimene (Limnos)
      Colour: on Catalan charts generally, and 15th-century Italian works in particular, this is almost always blue.

      Shape: see Colour & Shape Notes (Limnos)

37. metali (Lesbos)
      Colour: apart from the 1313 Vesconte atlas and two of the early unsigned Genoese works, this was left uncoloured until the 1367 Pizzigano chart, after which a variety of tints was used.

      Label: an uncoloured strip left for the island's name first appeared, with colour, in the anonymous Genoese atlas (Pujades A 9, assigned to the period 1325-50) and then on the 1367 Pizzigano chart, after which it was repeated on many Italian charts. Apart from two charts attributed to a Cresques atelier and Roselli's work, this device was not used by the Catalans.

38. sio (Chios)
      Label: a name slot was used on the 1330 Dalorto/Dulceti, the early anonymous Genoese atlas (Pujades A 9), on some charts assigned to a Cresques atelier and some Venetian works from the first half of the 15th century, as well as Roselli and Benincasa.

      Cross: my 1987 chapter in Volume 1 of The History of Cartography (p.378, note 66) mentioned five instances where Chios was highlighted by means of a red (Genoese) cross on an uncoloured background: the Vallseca charts of 1439 and 1447 (to which can now be added that of 1449), and three anonymous Venetian works: the so-called Pinelli-Walckenaer and Combitis (now Corbitis) atlases, and that in Lyons, Bib. Mun. MS 179 (Pujades A 28), which is assigned to the period 1425-50.
            It is not clear why the Genoese cross of St George should have been placed on those Catalan and Venetian works only, since the Genoese had controlled the island for most of the period after 1261. No later instance was noted in the 15th century but the cross was revived in the 16th century. Chios was lightly controlled by the Genoese until it was absorbed into the Ottoman Empire in 1566. However, as with other places lost to the Turks, chartmakers had a tendency to continue displaying supposed Christian ownership. The latest instance seen of the Chios cross was in 1673.

39. schiro (Skyros)
      Colour: this was invariably coloured red by the Catalans.

      Label: name labels were detected only on Venetian works: two anonymous atlases (Pinelli-Walckenaer and Pujades A 29), and then the 1470 Nicolo chart. However, it also appears on an undated Italian chart (La Roncière, et al., 1984) pl.24 [probably early 16th century].

      Shape: see Colour & Shape Notes (Skyros)

40. Rhodes
      Colour: Vesconte's first atlas (1313) coloured it alternating black and mauve (for decorative reasons) but left it uncoloured on later works. The convention of colouring it red was introduced by Dalorto/Dulceti and was followed by most thereafter, though some, e.g. Ziroldi, preferred gold.

      Cross: this was introduced by Dalorto/Dulceti in 1330 and was repeated on Catalan and some Italian charts thereafter, usually in the form of a white or black (silver?) cross over the red background. The island had been taken by the Knights Hospitaller in 1309 and would be lost to the Ottomans in 1522.

      Label: most Venetians (up to 1448) favoured a name label instead of a cross, although this was also first seen on the 1330 Dalorto/Dulceti chart as an additional feature (by placing the name along one axis of the cross). No later Catalan charts include the label.

Top of page

Black Sea
(to identify obscure features see the Visual index)

41-2. Danube delta
Presumably so as to emphasise the complexity and extent of the Danube's debouching channels, Vesconte initially showed seven 'islands', two of them larger and inland, in a selection of three colours. This is much the most noticeable feature of his earliest surviving work, the 1311 chart covering the eastern half of the usual region. From 1318 onwards he ceased to divide or colour the two former large islands, but used up to four colours on the smaller shapes.

Dalorto/Dulceti, in 1330, coloured a patch-work of twelve, similar-sized islands in four colours. A similar approach can be seen in two of the early unsigned Genoese charts and, intermittently, in later Italian work, though not on any Catalan charts. Thereafter (1339) Dulceti reduced the representation to a single, small coloured island - a form that would be much repeated from then onwards. There were two other approaches. First, Francesco Beccari in 1403, followed by Virga in 1409, gave separate colours (grey and red in each case) to a pair of islands, one large and the other small. This arrangement is also seen on an attributed Vallseca work and the two R. Soler charts. Second, a number of Italians (almost all Venetian) essentially ignored the Danube delta altogether.

To highlight the occasional difficulty of using this type of information for attributional purposes Ziroldi's five signed and four reliably attributed charts displayed three of the Danube delta types. His earliest work ignored it, the second and the latest three showed one small island only, and the middle four identified seven islands.

43-6. Dnipro estuary
43. The two small, thin islands running west-east [presumably the Tendrovskaya Kosa island and the long sandbank] were treated, jointly or singly, in too wide a range of colours for useful analysis, though a uniform blue was used on later Catalan work and by Benincasa.

44. The island to the east of that [Os. Dzharylgach?], in the bay set into Crimea's west coast, Karkinits'ka Zatoka, was usually coloured red, though sometimes blue on Venetian work or green on Catalan.

45-46. The estuary itself was usually represented by two islands oriented north-south, the larger one to the west labelled porto bo on Benincasa. This was either left uncoloured or frequently edged in a variety of colours. Given that the white space left in the centre of the 'island' would often be used for the name, it is sometimes difficult to judge if this constitutes a conscious name label. Such labels occur on some 15th-century charts: Venetian works starting with an anonymous chart assigned to the beginning of the century (Pujades C 28) and, among the Catalans, only R. Soler and Roselli.

On his 1311 chart Vesconte had broken into three the 'island' occupying the eastern part of the space, colouring the sections differently. Then, on both his 1318 atlases, he coloured gold an eastern island almost as large as the western one. The right-hand element was usually treated thereafter as a single entity and with similar inconsistency as to colouring. There was a brief Venetian variation comprising three distinct islands side by side, found on Briaticho (1430), Bianco (1436) and middle-period Ziroldi (1443-6).

47. Island marking the Enguri estuary, next to faxio / fasso
Half way down the east coast of the Black Sea the charts indicate an estuary, or rather they hint at its existence by showing a small inlet with an island at its head, beside the name faxio / fasso. This was often coloured red, presumably so that it would stand out in its isolated position. Occasionally, for example on the Catalan charts of Viladesters (1413-28) and Vallseca's of 1439 (but not, surprisingly, the Catalan/Estense world map), a large east-flowing river is shown at that point.

There are several, not particularly major, rivers emerging along that coast south of savastopoli (Sokhumi), where the Genoese were established until the mid-15th century. The chartmakers were clearly confused as to which river delta was involved and exactly where it was. Fasso, ancient Phasis Potamos, had immediately to its north two other names incorporating the Greek word for river, potamos. The largest of those rivers is the Enguri, which emerges, two names to the north of Fasso, beyond Lipotimo, at Negapotimo. I am grateful to Anton Gordyeyev for pointing out that the major estuary at this point does now (and did in the mid-17th century) have a triangular island dividing the river's bifurcated streams.

At least five different shapes for faxio / fasso can be identified, of which the 'spanner-head' (or catapault) was much the most common. However, since the island is small and at the extreme edge of most charts, where damage is likely, some of my interpretations have to be tentative and a number of instances could not be seen clearly. They are:

  • arrow-head: seen only on the 1456 Bertran & Ripoll chart
  • circle: (Italian) on both the earliest of the unsigned Genoese charts (Pujades C 4) and a slightly later one, as well as on most Venetian works; (Catalan) possibly on the 1330 Dalorto/Dulceti but rare thereafter [cf diamond and square]
  • diamond: (Italian) on the early Genoese atlas in Paris (A 9), on the 1409 Virga chart and three anonymous Venetian works, the latest thought to be c.1420; (Catalan) seen only on the signed R. Soler chart [cf circle and square]
  • spanner-head (old-fashioned type): (Italian) on the 1311 Vesconte chart and usual thereafter, e.g. Benincasa, and beyond the 16th century to at least 1630; (Catalan) first seen on the 1449 charts of Roselli and Vallseca
  • square: noticed just once on one of the unsigned Catalan works attributed to a Cresques atelier (C 19) but could easily be confused with the circle or diamond
Francesco Beccari's sole surviving chart (1403) uses the spanner-head form, as does one of the copies of his work in the Cornaro Atlas. Interestingly, the second Beccari copy shows the circle form. On those see Cornaro Atlas. The same page's Vallseca entry (towards the end) includes a discussion of the treatment of this feature on the signed and attributed Vallseca charts.

Top of page

48-50. Nile delta
(to identify obscure features and for an enlargement of Damiata see the Visual index)

48-50. Arrangement of the islands The last of the great river deltas depicted on the portolan charts concerns the area between Alexandria and Dumyat (Damietta). From Vesconte onwards, this was usually shown as one large triangular island with another (usually smaller) to its east. There were some variations to that pattern. Whatever the arrangement, the elements were often left uncoloured:

  • the smaller eastern island might be split in two: seen on the work of the Vescontes and on the earliest of the unsigned Genoese charts (C 4) only
  • the large western island could be split in two to form a middle island, sometimes coloured differently: examples are, (Italian) the Venetians - Ziroldi, Briaticho, Bianco and Nicolai only (i.e. 1426-48); (Catalan) the 1456 Roselli chart only
  • multiple islands (six): seen only on one of the early Genoese charts (C 9bis)
  • labels (three): on the Venetian atlas (A 30) only
51. Damiata
Finally, a very small but diagnostically useful observation. In a similar manner to the treatment of the 'a' of Aigues-Mortes (see 23 above), the first letter(s) of damiata was placed, invariably written in red, inside the eastern island. Dalorto/Dulceti was evidently the first to do this, separating out 'da'. As further evidence of the close contact between Majorca and nearby Genoa, three of the five early Genoese works (Library of Congress and Paris, BnF MS. Ital.1704 & MS Lat.4850) use this same convention. It might have come from Genoa, anyway, a supposition which is supported by the way that the earliest of the anonymous Genoese works (Florence, Ricc.3827) writes damiata as one continuous word but with the first two letters on the island and the currently unlocated Genoese chart sold at Hôtel Drouot (Pujades C9bis) places the separate 'd' and 'a' on different islands. {This paragraph altered 19 September 2012}

The Pizzigani, in 1367, were the first to place just the first letter separately, in a manner that was widely followed thereafter, from the Catalan Atlas onwards, on both Catalan and Italian charts. In many cases, the 'd' becomes a major feature framed within its own white box let into the island colour behind. As a pointer to continuing use of an old model, the Guillem Soler's 1385 reversion to Dalorto/Dulceti 's earlier 'da' form was repeated by his grandson (?) Rafel. The only other 15th-century instances of the 'da' form are seen on four unsigned Italian works assigned by Pujades to the period 1425-50.

Most of those who omitted this feature were, as might now be expected, Venetians - Virga, Ziroldi (apart from his middle period), Briaticho, Bianco and Nicolai. The latest instances noted were in the 1540s.

See How would the productions of an atelier be different from those of the master himself? [towards the end] for comments on the way that the treatment of the detached 'd' can apparently throw light on the status or training of a chart's author.

Top of page

Visual indexes to the Colour & Shape Analysis   |  Summary table of the features analysed (Microsoft Word doc.)

Colour & Shape Analysis Menu   |  Portolan Charts Main Menu

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

eXTReMe Tracker