E.1(A)
Paris, BnF, Rés Ge AA 567. Chart of the Aegean. 'c.1500' but probably 16th century. 73x40 cm.

Literature:
Astengo (2007) p.239 (FrP4) ['16th c.']
La Roncière & Mollat du Jourdin (1984), No. & Plate 24, p.214. The plate is captioned 'c.1500', but the catalogue entry, by Monique de La Roncière, says 'XVIe siècle'. She draws attention to the unusual four-circle arrangement of the rhumb lines and the coastal elevations. Their presence on the following chart leads her to consider both by the same chartmaker. It can additionally be noted that the name label over the Aegean island of Skyros has otherwise been seen only on 15th-century Venetian work up to 1470.

E.2(A)
Paris, BnF, Rés Ge D 7898. Chart of the coasts of Greece. 'c.1500' but probably 16th century. 60x40 cm.

Literature:
Astengo (2007) p.239 (FrP3) ['16th c.']
La Roncière & Mollat du Jourdin (1984), No. & Plate 23, pp.213-4. The plate is captioned 'c.1500', but the catalogue entry says 'XVIe siècle'. Attention is drawn, in the note written by Monique de La Roncière, to the incomplete rhumb network, which comprises just four lines (or the eight main directions, i.e. those normally drawn in black or brown). The north-west segment of the hidden circle (which would define the position of the 16 intersection points) is just visible. This is interpreted as evidence of an otherwise complete chart whose final stage of construction would have been the filling out of the remaining rhumb lines. For another example of such an incomplete network see E.13 and also the comment to E.11 (illustrated by Pujades (2007) p. 189) which I suggest might be an apprentice exercise. See also the note to the preceding entry.

E.3(A)
Munich, Bayerische Staatsbibliothek., Cod. icon 140 f.81. 49x79 cm.

Literature:
Kupčík (2000) pp. 106-9: Addendum 3 ('before 1500'). However, Rey Pastor & García Camarero (1960) p.193, had assigned it to 'Salvat de Pilestrina, 16th century', and Astengo (2007) p.242 [GeM4] also lists it as "[16th cent.]".

-- Munich, Bay. Staatsbib. Cod. icon. 138/40, f.82 [or Cod. icon 140 f.82]. See note to A9

E.4
Turin, Biblioteca Reale, Ms.O.XVI.5. Anon chart (late 15th/early 16th century?, an imitation of the work of Conte di Octomano Freducci?). 97.5x56.5 cm

Chet Van Duzer, in his BLJ article, 'Nautical Charts, Texts, and Transmission: The Case of Conte di Ottomano Freducci and Fra Mauro' (pp. 18-19), argues carefully against a direct attribution to Conte di Octomano Freducci. "While the Turin chart seems to show the influence of Freducci, there are too many differences between it and Freducci’s signed or securely attributed works for us to be able to assign it to him". This conclusion supercedes the comments that follow below. {This paragraph added 20 September 2017}

The chart covers the Mediterranean and Black Seas and extends from the British Isles to the Canaries. It has several compass roses, and town symbols with flags over Genoa (small) and Venice (large), presumably pointing to a Venetian origin. Those illustrations are similar to ones found in Benincasa's work (for example on the 1468 atlas in the British Library and on the 1482 chart). The treatment of the British Isles is close to Benincasa's, including the caption to Ireland's lacus fortunatus, referring to the 367 islands. But the Turin chart lacks what seems to be the Benincasa 'signature', the treatment of the headings for England and Ireland [see the Benincasa page under 'Headings for the British Isles' and Table 1]

The article on this chart by Corradino Astengo suggests that this is probably the work of Grazioso or Andrea Benincasa or of Freducci (see 'La produzione cartografica dei Benincasa e una carta nautica anonima conservata nella Biblioteca Reale di Torino', Bollettino della Società Geografica Italiana XI, 7 [nos 4-6?] (1990): 223-30). I noticed various features that seem to point to it being by Conte di Octomano Freducci rather than Benincasa. One is its treatment of the Red Sea. Omitted by Grazioso Benincasa on his two charts of 1461, this was included in his final chart of 1482 but with a different arrangement of the surrounding rivers to that found on this Turin chart. The outline here, however, is almost identical to that on the 1497 Freducci chart [see further on the Benincasa page under 'Red Sea'].

The Turin chart's handwriting does not seem to be by either of the Benincasas, but instead shows similarities to the 1497 Freducci chart [see the Benincasa page under Works by Benincasa's successors or by copyists].

Provenance:
This is not known

Literature:
Van Duzer (2017), pp.18-19
Astengo (1990c)
Terrae Cognitae. La cartografia delle collezioni sabaude. Catalogue of the exhibition in the Biblioteca Reale di Torino (19 November 2007 to 31 January 2008), edited by Maria Letizia Sebastiani & Clara Vitulo, with a portfolio of selected items (including this).

E.5
Keio University Library, 170X-24/1. Chart fragment (13x16 cm) extracted from a binding, covering the heel of Italy, Greece and Asia Minor.

Provenance:
Sotheby's 22 February 1972 Lot 537 ('c.1500') [see 'Census' 165]
Sotheby's 27 September 1990 Lot 509 ('first quarter 15th century')
Bernard Quaritch (London) Catalogue 1147 no.121 (Venice, 'c.1500')

Literature:
A brief description, in Japanese, by Takami Matsuda (p.447) and a reproduction on its front cover in: Eigo Seinen [The Rising Generation], published by Kenkyusha Co. Ltd. (October 2007).

Online scan in the HUmanities Media Interface (HUMI) at Keio University, 'European Illustrated Books and Manuscripts c.1400-1700'.

E.6
Rotterdam, Maritiem Museum, Atlas 66. Anonymous seven-sheet atlas, late 15th century, by a follower of Benincasa? 34 x 22.5 cm (atlas closed), each chart 34 x 43.5 cm.
The charts, each of which is glued to wood, are as follows:

1. Eastern Mediterranean and Black Sea
2. Central Mediterranean (Italy and Greece)
3. Western Mediterranean
4. North-west Africa, with the islands and part of Iberia [for a low resolution scan, enter atlas66 under 'Alle Velden', and click on 'Zoeken']
5. Atlantic coasts of Europe (British Isles to Spain)
6. Baltic Sea
7. West Africa (from C. Blanco as far as cauo de montte)

The Museum's brief catalogue description [enter atlas66 under 'Alle Velden', and click on 'Zoeken'] assigns the work to Grazioso Benincasa, with the date 1450, but it also states 'Vermoedelijk omstreeks 1470-1480, vervaardigd door Grazioso Benincasa' [probably about 1470-1480, produced by Gazioso Benincasa]. The similarities with the work of Benincasa are certainly striking - for example the diagonal scales across the corners. The precise treatment of these suggests that the model used here may have been one by Grazioso's son Andrea. However, the scale edging and the sheets' outer borders incorporate more lines than usual in Benincasa atlases [for details of these see the Microsoft Word format Benincasa: Table 3].

The typical three-line note is present across the middle of Ireland but, as described for the anonymous Turin chart [E.4], sheet 6 in this atlas lacks the distinctive treatment by Benincasa of the headings for the separate countries of the British Isles. In addition, instead of the invariable Scocia, the form here is schozia. The headings are also written in a different way to those on the Turin chart. Confirmation that the Rotterdam atlas is in a different hand, both to that of the two Benincasas (Grazioso and Andrea) and to the draftsman of the Turin chart, can be seen, for example, in its characteristic use of an open 'a', which appears like an 'n' [on which see the Benincasa page under Works formerly attributed to Benincasa].

The 'Colour & Shape Analysis', in which the entry for this atlas appears on the Microsoft Word table, Colour and shape on the productions of Benincasa's successors and imitators, concludes that 'if this atlas was based on a Benincasa original, the unidentified copyist was not concerned to imitate minor stylistic details'.

The most distinctive features of this atlas are found in the final two sheets, devoted respectively to the Baltic and West Africa. Sheet 6, covers the Atlantic coast from Bruges up to Norway, and then into an unrecognisable Baltic. Although the outlines are highly traditional, and are found on two Benincasa charts - the signed work of 1482 in Bologna and the chart in Florence (Archivio di Stato, C.N.9 - Census 71), whose date and signature seem to have been removed - there is no equivalent sheet in Benincasa's surviving atlases. That said, there is a precedent for Benincasan atlases with additional sheets. For example, the seven-sheet atlas now in private hands (formerly H.P. Kraus - Census 179) placed Ireland on a separate sheet and the now untraceable 1472 atlas formerly with Luigi Bossi (Census 172) had eight sheets, as does the work in Paris by a copyist (BnF Ital.1698 - Census 29).

The West African coverage includes the extra sheet introduced by Benincasa in 1468, which provides the earliest possible date for this atlas. The dividing point for the three sheets that take in the west African coast is not the same as that found in those surviving atlases for which this information is available [see Benincasa: Table 2]. However, the distinctive abbreviation for san vicenso (san v.co), at a point where space was not restricted, confirms its dependence (even if indirectly) on a Benincasan model [see Benincasa: Table 4].

For further comments on the treatment of these and other features of this atlas compared to Benincasa's signed work, as well as a discussion of the handwriting of Grazioso, Andrea and their successors and imitators, see here.

Provenance:
Sjoerd de Meer has kindly supplied the following notes, with English translations:

1. From the F.C. Wieder papers in the Maritiem Museum. "nr 161 Venetiaansche Zeeatlas der xve eeuw, type Benincasa. Met 7 kaarten van de kusten van Europa en het Noorden van Afrika waarvan in het bizonder vermelding verdient een kaart van de Oostzee. In kleuren geteekend op perkament in een gelijktijdigen Venetiaanschen band in bruin marokijn met blinde en vergulde stempels". [nr 161 Venetian sea-atlas of the 15th century, Benincasa-type with 7 charts of the coasts of Europe and North Africa, and also a chart of the Baltic. In colours on parchment in a contemporary Venetian binding with blind and gilt stamps.] Added in pencil is 'WAER' [i.e. Willem Anton Engelbrecht Rotterdam].

2. F.C. Wieder, 'Uit de grooten tijd der Oud-Nederlandsche cartographie', in: Tijdschrift van het Koninklijk Nederlandsch Aardrijkskundig genootschap (1916): 554-68, particularly pp. 566-7.

"Ik trof aan een dergelijken zee-atlas van het type Benincasa, afgebeeld bij Nordenskiold, Periplus. Het is een kleine folio-band, bevattende zeven op perkament geteekende kaarten, waarvan zes te vergelijken zijn met correspondeerende kaarten in Benincasa-atlassen. De zevende is een elders mij niet voorgekomen kaart van de oost-Zee, doch alleen het Zuidelijk gedeelte van een primitieve voorstelling, maar waarop de beroemde middeleeuwsche stad Wisby zeer duidelijk is aangegeven. De overige kaarten stellen de kusten van de Middellandsche Zee, West-Europa en het Noord-Oosten van Afrika voor. De atlas is gebonden in een verguld lederen venetiaanschen band uit den tijd, is zeer fraai van uitvoering en uitnemend bewaard. Hij dagteekent uit het begin der XVIde eeuw of wellicht nog uit de XVde." [I found an atlas of the Benincasa type, depicted in Nordenskiöld's Periplus (*). It is a small folio-binding containing 7 charts drawn on vellum. Six can be compared with Benincas atlases. The seventh is a chart of the Baltic not known to me. It is only the southern part, primitively drawn, but with the famous medieval town of Wisby clearly given. The other charts are of the coasts of the Mediterranean, Eastern-Europe and the North-East [i.e. West] of Africa. The atlas is bound in a contemporary Venetian gilt leather binding and is splendid in design and well preserved. It dates from the beginning of the 16th century or perhaps from the 15th.]

   (*) a reference [not, as stated, an illustration] (in Periplus p.60b) to the only other seven-chart Benincasa (or Benincasa-type) atlas known, then (1897) in the possession of the Princes of Lanza di Trabia (Census 179). It could not be the Rotterdam atlas, since Census 179 was still with Prince Trabia e Butera in 1950, later passing to the New York dealer H.P. Kraus and into an English private collection.

3. "In 1924 some gentlemen of Rotterdam, Ruys, Engelbrecht, Goudriaan and Van Es (ship-owners and brokers) gave the atlas in loan as part of an extensive collection of globes and atlases. It was never researched how they acquired this collection. Part of the collection was the "zevenbladige Venetiaanse zee-atlas geteekend op perkament, pl. m. Anno 1500" (seven-sheet Venetian sea-atlas drawn on vellum, c. 1500). In 1950 the Museum bought the collection, which was known as the REGE collection (after the four owners)."

Sjoerd de Meer therefore suggests that F.C. Wieder bought the atlas about 1916 and sold it to Engelbrecht, who included it in the REGE collection in 1924.

Literature:
Mentioned on p.13 of the January 2002 Newsletter of the Brussels International Map Collectors' Circle (BIMCC) as 'a most intriguing anonymous Venetian portolan atlas of around 1500 which appears not to have been studied as yet'.

[Thanks are due to Corradino Astengo and Sjoerd de Meer for help with this entry; the opinions though are my own.]

E.7
Oslo and London, Schøyen Collection MS 1981. Fragment covering Apulia [originally part of a chart of the Adriatic?], recovered from the binding of a religious manuscript. 23x17 cm.
'Portolan map of the Adriatic coast of Apulia in the kingdom of Napoli from Rinaldo south as far as Port Badisco, including Otranto and inland towns with little hills'. The small section of the Italian coast covered, and the suggestion from the fragment's size that it might represent about one-fifth of a complete chart, points to it being part of a chart of the Adriatic. The only illustration identified is that in the Sotheby's sale catalogue.

'Binding: Napoli, Italy, 16th century, limp vellum, sewn on 3 bands, formed of 'late 15th century portolan map', 23x17 cm'. In a manuscript of the first half of the 16th century containing religious works by Aquinas and others, written by a nun.

Provenance:
1. Benedictine monastery of Santa Patrizia, Napoli (16th c. -)
2. Private collector, Connecticut (-1994)
3. Sotheby's 5. 12.1994:88. [= lot 88 on either 12 May or 5 December 1994]

E.8(A)
Lisbon, Torre do Tombo. Two sections (almost adjoining) covering most of the Mediterranean and Black seas, with a missing strip down the middle, loss at the extreme west end of the Mediterranean, and slight trimming on the other sides. Garcia gives the size as 49x61 cm.

It was originally thought to be late 15th or early 16th century. Tesouros p.70 notes the Spanish arms over Mellila (conquered in 1497) but not Oran (conquered in 1509). However, evidently on stylistic grounds, Astengo considers it to be a Catalan chart, possibly as late as the 17th century. The latest attribution, by Joaquim Alves Gaspar in June 2021, is to Jaume Olives, c.1550-60 [see the MEDEA link below for the evidence behind that attribution, in a note written by Gaspar and Sima Krtalic].

Provenance:
Santa Casa Misericórdia de Cabeço de Vide, southern Portugal [deposited in Torre do Tombo]

Literature:
Gaspar & Krtalic (2021)
Astengo (2007) p.251 [PL2], note 43
Garcia (1990)
Pinheiro Marques (1993) (illustrated, p.43 top right, the left-hand section)
Tesouros da cartografia portuguesa (1997), p.68 (the two sections illustrated), 69-70 (with further references), 118 (no.21)

See for online illustration, the MEDEA-CHART Database.

[Compare with Census 126 describing another fragmentary chart (in this case a single square section), also now considered to be post-1500.]

Viana do Castelo, Arquivo e Memória do Município de Viana do Castelo, 5.A5.3.1 and 5.A5.3.2. Adjacent fragments of a single chart (ex separte bindings), late 15th/early 16th century, comprising, first, a vertical strip with the Atlantic coasts of Ireland, Iberia and north-west Africa (51x26cm) and, second, the western Mediterranean (48x27cm).

Literature:
José Manuel Garcia, 'Uma nova carta de marear: notícia do seu descobrimento'. Oceanos 5 (Lisbon, 1990): 25-8 with full-size colour illustration
Pinheiro Marques (1993) (illustrated, pp 42-3)
Pinheiro Marques (1989a), p.319
Tesouros da cartografia portuguesa (1997), pp.70-71 (illustrations, 70-71, 118-119 (nos 22-23) ('probably early 16th century')
See for online illustrations, the MEDEA-CHART Database.

E.11
Barcelona, Arxiu Històric de la Ciutat de Barcelona, Gràfics. Fragments of a blank chart, with four scale bars, assigned to the first half of the 15th century. 54x84 cm.
Removed from a binding of a protocol of the notary Joan Reniu of 1430 (on whom see Pujades, 2007, note 6 (p.483)). In the absence of handwriting, Pujades considers that the style of the graphic scales fits with a similar date (private communication).

The fragment contains what might be considered to be the first stage of construction only, i.e. the mathematical sub-structure. In this case the rhumb network comprises two (unusually small) contiguous circles. However, there are two problems with that interpretation. First, the scale bars would surely not have been placed at that early stage, since, while the coastlines and names could be written anywhere on the prepared sheet, no chartmaker would have allowed a previously-placed scale to interfere with the chart's later outlines. The second factor is that there appears to be a very small, faint section of coastline, in the lower left of Pujades's illustration. Are we seeing here - as has been suggested also for E.2(A) and E.13 - an apprentice's exercise?

Pujades (note 19, p.483) suggests the following: 'It is very hard to say, given the fragment's poor state of conservation, whether or not work had begun on drawing the coastline, an issue that has its importance since, if the small strokes we perceive may not be regarded as sectors of the coast, we might be led to believe we are in the presence of an experiment in the execution of a wind network'.

Literature:
Pujades (2009), illustrated p.51
Pujades (2007) p. 67 (C 55), pp. 472b, 473b, and notes 4-6, 19 (p.483), illustrated p.189.

Literature:
Luisa Martín-Merás, Carlos V. La Náutica y la Navegación (Madrid: Sociedad Estatal para la conmemoración de los centenarios de Felipe II y Carlos V, 2001), No.20 'Carta de navegar para la derrota desde la barra de Sanlúcar a las Islas Canarias, Juan Excalante de Mondoza, c.1550'.

The fragment shows the coast of France with clearly visible names, and likewise a section of the north African coast. There are several strange features about this chart. Southern Britain and Ireland are included, but appear faintly outlined without any visible lettering. Presumably, the intention would have been to include the whole Iberian coast, with the Balearics, and the south coast of France, but they are invisible. Rossello Verger notes that the chart was removed from the binding of a book of protocols (without indicating its date). He suggests it is a page from an unfinished atlas and points out that 'the oblique grid (67.5x88 mm) forms part of an unfinished martelory, which must have been used for transcription'. However, there is no convincing evidence that the network played a part in the plotting of portolan chart coastlines.

The rhumb line network (whose 'hidden circle' can be seen passing through Ireland) does appear to have just the eight main wind directions (those that would normally have been drawn in black or brown). It is possible that the other colours had faded, but there is a precedent for such a partial network in the Paris chart E.2(A). Indeed Pujades (2007), p.474a provides examples of charts in which there was a 'system of first drawing only the lines that appear on all charts in sepia/black, then outlining the coasts, subsequently copying the black toponyms and finally drawing the remaining lines (green and red) of the wind network'. He notes that on the 1339 Dulceti and 1439 Vallseca charts (Census 13 & 128), as well as some later ones, 'while the part featuring the sepia lines (representing the directions of the eight major winds) stretches from one end of the chart to the other, the complete wind network leaves out the land not contained within the circle, which indicates that the order of execution was the same as in the case of the chart preserved at the Arxiu del Regne de Mallorca'.

The chart has certainly suffered from its later use but the fact that the French coastline (with its clear black lettering) appears no stronger than that of Britain (without any visible lettering at all), prompts the suggestion that this might have been the work of an apprentice under training, thus supporting Rossello Verger's suggestion that it is unfinished. If so, it would presumably have been abandoned rather earlier than a finished chart, and perhaps placed in the binding closer to its date of construction. For which reason, it is unfortunate that the date of the protocols is not available. [See also comment to E.11.]

Literature:
Ginard Bujosa (2006) pp.119-20
Pujades (2007) p.65 (C 76), 'second half 15th century', pp. 473b-474a, note 22 (p.483), illustrated p.189
Rossello Verger (1995) No.1 (pp.63, 361) 'its round Gothic script dates it to around 1400'.

Provenance:
Conde de Zavellá
Gabinete de Antigüedades de los Capuchinos de Mallorca [this sequence is not clear]

Literature:
Astengo (2007) p.252 [SpP1]
Exposicion de cartografia mallorquina (1990) 16 (page or number)
Ginard Bujosa (2006) p.118
Montaner (1976)
Rosselló Verger (1995) Pl.15 "1500-1600"

E.15

Sidney R Knafel Collection. Unsigned 15th century chart, attributed to a Beccari workshop. 66x117(or 120) cm
It includes Antillia and the other fictitious Atlantic Islands. Christie's noted similarities with the depictions on the Pareto and Canepa charts (1455 & 1489) but notes that this includes 'Insulle de novo reperto' to the west of Antillia. The vignette of Genoa is much the largest urban depiction.

Provenance:
'Mr. Robert du Celier(?) Conseillier' - late 16th century ownership inscription
Counts of Maldeghem, Bruges - Christie's refers to Philip of Maldeghem (1547-1611). 'His son Robert (d.1654) was a member of the council of war of the Infanta Isabella of Spain, but later in life assembled and put in order the large family archive, and it is almost certain that the charts [this was one of a group] were then already in his library'. The collection passed to the present owner by descent
Christie's 13 April 1988 Lot 172 'Genoa, 1460-80' [Barnes]
Christie's 20 June 1990 Lot 43 'Genoa, second quarter 15th century', attributed to Battista Beccari [Martayan Lan]
Sidney R Knafel Collection [the map collection was donated in 2011 to the Phillips Academy, Andover, Massachesetts and the items are being digitised (2021) by the Norman B. Leventhal Map & Education Center. The donor died in 2021; this chart does not seem to have been part of that donation, hence its current location is not known.]

Literature:
Alfredo Pinheiro Marques (apparently unpublished paper distributed at the International Conference on the History of Cartography (Amsterdam, 1989))
Suárez (1992) pp.30-6, plates 4-6 - attributed to Battista Beccari, c.1434 [reason not explained, though possibly related to the dates on the two signed B.Beccari charts, namely 1426 and 1435]
Pujades (2007) p.67 (C 53) - 'second quarter 15th century, Genoese, workshop of B. Beccari'

E.16

[Bruce Ferrini, 1997]. A six-sheet atlas, each 24 cm square, estimated by Sotheby's as c.1400 but perhaps second quarter 15th century.
Sotheby's claimed that this 'is almost certainly the oldest European atlas in private hands, and it is the oldest complete atlas offered for public sale since 1897'. Their description notes Venetian usages, e.g., 'Zenoa' and 'Lizorno', and sees similarities with the work of the Vescontes, whose work, they suggest, continued to be copied into the 15th century. The naming of Livorno, which the Chapter (p.427a) noted was first observed on a dated chart in 1426, casts doubt on the suggested dating. However, the omission of Cherbourg (first noted in 1318) and Harfleur (1385) - see the same reference - need not, as Sotheby's claims, count against a date well into the 15th century. The failure to show fresh information may be the sign of an outdated model or, perhaps in the case of small atlases sheets such as this, the omission of some names could be due to lack of space. But the inclusion of a 'new' name cannot be readily ignored.

{Update note 3 September 2016}
Jens Finke kindly drew my attention to online scans of this atlas, conveniently arranged on a single sheet. However, they are of low quality and the names remain largely illegible even when using 'superzoom' [the uncredited images are on an undated page of the fine art reproduction company Bridgeman, Berlin]. The style is close to that of Pietro Vesconte although the images are not clear enough to compare the hands. [Unaccountably, Kunstkopie attributed the atlas to Grazioso Benincasa (fl.1461-82), whose own distinctive style is very different to this.]

The sheet featuring the British Isles is diagnostically the most useful when considering supposedly early charts. In this case, the outline and toponymy match the stage reached on the Vesconte atlases by 1318. Whereas all the British toponyms were in black on the 1313 atlas, the handful that appeared in red in 1318 are shown thus here. Likewise, the Ferrini atlas includes the four names introduced by 1318 between Portland (porlan) and Southampton (antona ). The Ferrini atlas does not depict Ireland or any of the other additions found on Vescontian works from c.1320-c.1330. There is no obvious reason - apart, perhaps, from carelessness - for the omission of the prominent London (written inland) and the usual three west coast names. [On the British Isles development see Carte Pisane Hydrography Tables – Table 1 ‘Development in the outline and toponymy of the British Isles’, and Table 2 ‘Early names along the South coast of England’.]

No conclusive verdict is possible without access to fully legible scans. The work certainly looks superficially Vescontian. The way that the Galician toponymy runs awkwardly into the outer border might suggest the work of a less competent copyist. Except that the same thing can be seen on the 1318 atlas now in Venice. However, there are at least three significant stylistic differences between this atlas and the signed survivors. One is that Vesconte's habitual border comprises chevrons of alternating colours whereas the Ferrini's segments are plain rectangles. The second concerns the dotted letter 'i', a form evidently not seen until much later. The other anachronism relates to the convention of leaving uncoloured rectangular panels (labels) within a coloured island into which toponyms could be written. These can be seen here on the gold-tinted Crete and Cyprus but that is a feature otherwise first noted on the 1330 Dalorto/Dulceti chart and was not one used by the Vescontes [Name labels.]

Taking note of the way that this atlas reproduces Vesconte's very distinctive initial letter for the England heading (Inglitera), it seems likely that this represents an unauthorised imitation of a Vescontian atlas. Given the notable additions and improvements made by the Vescontes in the 1320s, the model available to a copyist must certainly date from before that time. The presence of Livorno, and indeed the Venetian spellings mentioned in the auction catalogue - if confirmed - would certainly throw further doubt on both a Vescontian authorship and a 14th-century date. But the recently discovered (and, in my view, early) Lucca chart also includes Livorno and I have argued elsewhere that the significance of this toponym may have been overstated [Carte Pisane essay (Livorno)].

Provenance:
Christie's 20 June 1990 Lot 44 ('Venice, last quarter of the 14th century' - illustrating the Adriatic sheet]

Sotheby's 17 June 1997 Lot 64 ('North-east Italy (probably Venice), c.1400 - illustrating the Adriatic sheet).

A catalogue issued by Dr Jörn Günther (Hamburg) and Bruce P. Ferrini (Akron, Ohio) in Autumn 1997, 'Recent acquisitions: medieval & renaissance illuminated manuscripts' (illustrating the Adriatic sheet) [see Robert W. Karrow in The Portolan 75 (Fall 2009) p.51 - supplemented by personal communication].

Bruce Ferrini, 1997 - the collector/dealer Bruce Ferrini had his collection seized in 2005 and sold in 2008. Most were biblical items and the publicity focussed on those. The fate of the atlas is unknown. See 'Akron investor's biblical artifacts being auctioned online' by David Giffels in the Akron Beacon Journal, 14 March 2008. This describes the online auction then underway, involving 153 of Ferrini's items: 'the sale is being conducted by the Dallas-based Heritage Auction Galleries, with the actual items on view at Arte Primitivo Gallery in New York City. Online bidding has begun and will continue until the close of business Wednesday [i.e. 19 March 2008]'. While the Heritage Auction Galleries and Arte Primitivo websites includes past sales, there was no sign of the Ferrini material (in November 2008 - though the Heritage Auction Galleries requires registration).

Information was received from Dr Jörn Günther in November 2021 that the atlas had been sold to a private collector in Switzerland. It is not clear just when that happened.

Low-resolution scans are available in the relevant Medea- Chart record.

E.17 [The 'A' has now been removed since the chart's date has been confirmed as c.1360, and it was likely drawn in Venice. The new owners have dubbed it the 'Rex Tholomeus' portolan chart because of the full-length portrait of Ptolemy. This was formerly included in the 1986 Census as No.167, 'mid-14th century']

Barry Lawrence Ruderman Antique Maps Inc. (2022)

This is a duplicated entry for a mid-14th century chart, originally with Prince Filippo Corsini in Florence, which was no 167 in the original Census, since at that time its location was unknown. It has now been re-identified by the present owners, who have produced a 49-page catalogue: 'The Rex Tholomeus portolan chart of 1360', authored by Barry Ruderman, Alex Clausen & Hamilton Allport. It is unlikely that any portolan chart has been the beneficiary of such a detailed analysis. There is also a summary description, entitled. 'The Oldest Portolan Chart in America and The Fourth Oldest Surviving "Complete" Portolan Chart of Europe'.

A carbon-14 analysis pointed to a bimodal time period of 1324–1355 or 1392–1437. To narrow that down, the authors noted that 'Numerous cities that underwent shifts in governance between 1370 and the mid-15th century, such as Avignon, Bordeaux, and Pisa, all display flags representative of the earlier era.'

Among a number of interesting findings is the comment that 'early renditions of the British Isles usually show sparse detail on the west coast of Scotland and on the northern coast of Ireland. That is not the case here; both regions are dense with toponyms that appear on no other charts, even on those dating from the fifteenth and sixteenth centuries.' Another reference is to 'an unusually rich rendering of Norway depicts the gyrfalcons that so captivated medieval nobles.'

The chart is easily recognizable, first, by the westwards extension, necessary to accommodate part of the Canary Islands, second - a unique feature that confirms its appearance in at least three Christie's sales - by the area of western Europe that has been scraped away and, third, by the topographical drawing placed over southern Scandinavia.

Provenance: First noted in 1888 in the possession of Prince Corsini in Florence, it is next recorded with John Alfred Spranger (d. 1968) and was consigned to Christie's on 12 May 1993 by his daughter Elizabeth Graff (as circa 1500) and acquired by Ann and Gordon Getty, then consigned to Christie's again on 12 October 2022 in the Getty sale (where it had a suggested date of 1530).

For scans see the full and summary reports above, and also the Medea-Chart Database.

Updated: 6 February and 18 May 2023.

E.18

Hôtel Drouot, Paris. Chart by ['Perrinus Vesconte', c.1320-25']. 43x73[or 76] cm. [see Census 177]
Removed from a binding (Ridolfi family), cropped and with holes. For doubt on the attribution to Vesconte see Chapter p.407, n.274 & p.378, n.67. The toponymic analysis in Chapter pp.416-17, where it is no.12, demonstrated statistically that it is not a Vesconte work. Featured in 1990 'Portugal Brazil' exhibition as 'anonymous owner'. See also Anonymous works and the question of their attribution (No.6).

Provenance:
Dudley Barnes 1977 (acquired from Franco Novacco)
Sotheby's 15 April 1980 Lot A
Israel Catalogue 22 (1980) no.1
H.P.Kraus Catalogue 178 (1987?) no.187
Hôtel Drouot 30 March 1992 Lot A

Literature:
James E. Kelley, On Old Nautical Charts and Sailing Directions: Technical Essays (Melrose Park, PA: Sometime Publishers, 1999), pp.114-20 [but the ‘draft’ is dated 1977, the year Kelley was shown it] – this reference was previously overlooked [5 March 2020]. Despite subsequent claims, Kelley had concluded that, although it was likely to date from the second quarter of the 14th century, the chart was not by either Vesconte.
Pujades p.65 (C9 bis) [Anon. Genoese, c. 1325-50], see p.447, note 5 (where he doubts the attribution but is unable to read the northern Adriatic names from the available reproduction)
Portugal Brazil: the Age of Atlantic Discoveries (1990), p.33

See also Census, No.177.

E.19

[H.P. Kraus, 1989]. Six charts on a single sheet, 35.5x59 cm (each 14.3 cm square). Venetian (?), c.1460-70.
This unusual work is very close, in format and content, to another 'flat atlas', the 1462 Nicolo Fiorino sheets in Vienna ['Census' 2 and the Pujades scan (A 32)]. This similarity has not apparently been noted before. The present work comprises six charts on a single sheet, similar in format to a portolan chart and, like them, with a neck at one side. Fiorino's production takes instead the form of two sheets, each containing four small charts. Since each sheet has its own neck, it is evident that they were always separate.

Although the divisions between the separate charts and the coastal outlines are not a perfect match between the two works, they are remarkably close. Both provide separate charts for the east Mediterranean, Black Sea, Aegean, Adriatic and a pair for the central Mediterranean, including, on the one hand, Sicily and, on the other, Corsica, Sardinia and the Balearic Islands. Fiorino is alone is extending to the west, as would have been expected, on two further charts that cover the western Mediterranean and then the Atlantic coasts up to the southern half of the British Isles. The truncation here would have made no practical sense and the plausible suggestion by the Christie's cataloguer in this case that the stitching holes at that end of the sheet might have attached two further charts is clearly borne out by comparison with the Fiorino sheets. The blank shield at the neck, with its surviving vellum tie, all indicate a life spent as a single, rolled-up chart.

Fiorino was a Venetian shipmaster (see Falchetta (1995) pp.60-2) and the Kraus sheet was considered a Venetian work. It is tempting to attribute the Kraus sheet to Fiorino, and the suggested date, 1460-70, matches precisely Fiorino's dated work. However, even making allowances for the more cursive hand on the Kraus charts compared to Fiorino's neat writing (and working from reproductions of the two works that are far from clear) there do not seem to be any distinctive similarities in the hands. The regular use of the 'z' letter form on the Kraus sheet, which seems not to have been used by Fiorino, and different application of long tail flourishes - on a 'g' by the Kraus's author and on an 'h' by Fiorino - point to these being the work of two different people.

Nevertheless, they are the only surviving works of that nature, are both thought to be Venetian, and are apparently from the same period. If not copied from one another, they must have shared a common model. A further, significant, linking element is that in neither case were the sheets intended to be cut up to form an atlas. Some details run across the dividing borders - a flag between the charts of southern Italy and Greece, and the Genoa vignette, for example on the Kraus sheet, and a number of place-names on the Fiorino sheets.

Other features were noted by Christie's. Most interesting is the 'apparent indication of local scales. Each inset shows the local miles i.e. mila davenezia and mila de larzipelago'. There is no sign of such an indication on the Fiorino.

In 1989, when this sheet came up for sale, the comparison with Fiorino was not made. That was understandable given that no reproduction apparently then existed. Instead, Christie's drew attention to apparent similarities with the small bound collection known as the 'Luxoro Atlas' (Census 81). This is now datable to before 1421 according to Pujades (2007) p.69 (his A 16), and the hand is clearly different from the Kraus or Fiorino works.

Despite the considerable time gap between the Luxoro Atlas and the Kraus and Fiorino sheets, the comparison is a valid one. The eight charts of the Luxoro are arranged as four facing, and unconnected, pairs, following very similar divisions of the coastlines to those found on the later works. The Kraus and Fiorino arrangements can best be seen as an extension of the format adopted by the Luxoro's Venetian author, Francesco Cesanis, into a type of flattened-out atlas, that seems to have been unique to that city. Pujades mistakenly gives the sheet size of the Luxoro Atlas as 16x23cm (being misled by an enlarged reproduction) instead of the correct 11x15 cm. These are therefore the smallest sheets recorded for a pre-1470 atlas and the Christie's sheets are not much larger. Significantly, the Luxoro Atlas extends to the western edge of the Mediterranean and then up to southern England - like the Fiorino sheets and, as has been suggested above, like the probable original format of the work described here.

Since the sheets of the Luxoro Atlas are not square (as are those of the Kraus sheet, or with their width arranged to suit their subject as are Fiorino's) its author might have found some cardinal points more convenient than others. Instead he uses all four directions, thus demonstrating the irrelevance of consistency in that respect to the users of such navigational aids. The Kraus and Fiorino sheets, on the other hand, offer a constant southerly orientation for the four eastern sheets, though they differ in the central Mediterranean. The network of 32 rhumb lines (a feature first found on Roselli's 1456 chart, and present on both the Fiorino and Kraus sheets) suggests a date not earlier than that.

A thorough comparative analysis of the three works discussed above must wait for reproductions of adequate quality, and those that allow direct comparison. If the unknown owner of the ex-Kraus sheet sees this, perhaps they might consider assisting in this way. In the meantime I would like, once again, to acknowledge the public-spirited foresight shown by Ramon Pujades in bringing this material, including the Luxoro Atlas and Fiorino sheets, within reach of researchers worldwide.

Provenance:
Christie's 21 June 1989 Lot 40 ('Venice, ca.1460-70', perhaps drawn by a Catalan) [Kraus]

E.20

Sotheby's 6 December 1988, 'Western Manuscripts and Miniatures', Lot 37. Portolan chart signed and dated by Petrus Roselli, 1469. 66x111 cm.
It is described as having 'emblazoned shields of many countries, nine town views, four wind discs, five distance scales...'.

Provenance:
Harrison D. Horblit (?)
Cologne, Museum Ludwig (Dr Peter Ludwig of Aachen), Ludwig MS XIII.14
J. Paul Getty Museum, Malibu (acquired 1983)
Sotheby's 6 December 1988, Lot 37 [ Schiller] "a small part of the great Sammlung Ludwig, the most outstanding library of medieval manuscripts assembled in the last half century. The collection was formed (mostly through the agency of Mr. H.P. Kraus) by Dr.Peter Ludwig, of Aachen, and was at one time on deposit at the Schnütgen-Museum in Cologne. In early 1983 the entire Ludwig manuscript collection, comprising nearly 150 items, was sold en bloc to the Getty Museum. The lots are described in the order of their listing in Anton von Euw and Joachim M. Plotzek, Die Handschriften der Sammlung Ludwig, I-ÏV, Cologne, 1979-1985" [in Vol. III (1982), pp.302-4, Plate XIII,14]. See also 'Acquisitions, Supplement: Manuscripts', The J. Paul Getty Museum XII (1984) p.303, no.111.

It is highly probable that this chart - the only signed and dated work to surface (and then disappear again) since 1986 [but see now, May 2015, E.27] - was that formerly owned by Otto Heinrich Friederich Vollbehr (b.1869) of Washington, D.C. [Census 180]. It was signed and dated by Roselli from Majorca. De Ricci (1935, p.504, no. 15379) noted that, in that year, the chart was on loan to the Library of Congress. He described it as covering the Mediterranean, France and the British Isles, and measuring almost exactly the same as the Ludwig chart (112x67 cm). I am not aware of any illustration or description of the Vollbehr chart.

E.21

Jesi [or Iesi], Biblioteca Comunale, Palazzo della Signoria, Direzione [MS number not recorded]. The north-east quarter of an Italian chart recovered from a binding, covering Italy, Adriatic, Aegean, Black Sea and Asia Minor, 36x42 cm.

The chart's rhumb centre is in south-west Italy, which implies that it would have extended as far west as the British Isles and Canary Islands. The chart is most unusual (at least for the mid-15th-century date claimed) in having inland detail, evidently in the same hand. This includes province names, stretching from Moravia to Tartaria and Armenia M[aior], along with rivers, mountains and inland town symbols. Apart from some vignettes in central and eastern Europe, and two flaming [Roman] altars north of the Black Sea, most of the settlement detail is found in Asia Minor.

This partial chart was first noted by archivists at the Biblioteca Planettiana of Jesi and communicated to Francesco Bonasera in 1985. He provided a brief description and listed the provincial names. His only evaluative comment was: 'Come epoca, dal carattere delle scritture, si può ritenere che sia relativamente tarda, cioè del XVI secolo'. This tentative dating to the 16th century was followed by Corradino Astengo (2007, p.245 - ItJ1). Bonasera's 1997 article largely repeated that of 1985, while adding a good quality illustration (Tav. XXXVI).

Patrizia Licini, in a succession of articles since 1991, based on close observation of the original, has argued for placing the chart in the second half of the 15th century. She also noted that Trebizond 'the last bastion of the Byzantine Empire, which fell in 1461, is still resisting on the Jesi map' (2000, p.63). However, it is not unusual to find chartmakers refusing to admit Christian defeats and the 1497 Freducci chart, for example, continues to show the Byzantine arms over that city. It is not clear what evidence there is on the Jesi chart that the Empire of Trebizond (Trabzon) has yet to fall to the Ottomans. Likewise, the occasional crescents that can just be made out on top of the towers of one or two of the town vignettes in Asia Minor provide a terminus post quem for the 1460s not confirmation for such a date.

Licini interpreted various German town vignettes as a record of a journey made to eastern Europe in 1452 by Aeneus Silvius Piccolomini. That directly related, in her view, to the main purpose of the chart/map, namely as 'a visual crusading sermon', or a 'manifesto', created in the context of the attempt in 1464 by that same man, now Pope Pius II, to mount a crusade (2000, pp.63-64; 2008, pp.208-210, 218). What she describes as 'dismantled' towers, indicating post-conquest Ottoman destruction, are found widely across Asia Minor. An alternative interpretation would be that these were left 'unfinished' by the map maker, as seems to apply to the larger symbols for Paderborn and Grabau. The simple, upturned table effect of the majority can be seen to form the drafting basis of the vignette construction for the few that were completed. Perhaps these, with their names, were placed there as markers, for the future elaboration that never took place. I could not find clear confirmation of an attempt to convey structural damage.

Licini's 2008 article (based on a 2006 conference paper) is almost entirely concerned with the map , rather than chart aspects of the Jesi fragment, adding for example a discursive passage about the possible influence of Nicholas of Cusa, as a cosmographer, on the work's construction. She sums up her aim as being 'to prove that the Iesi fragment offers a geopolitical plan through specific key cultural concepts expressed in the form of views of castles and in place names that lie at the heart of its strategic design. Select place names and views of cities point to Aeneas Sylvius Piccolomini as the main character in the Iesi portolan chart's story' (p.208).

As to the chart's authorship, Licini, settled on the Freduccis, a powerful family of counts in Fermo near Jesi, both of which towns formed part of the Marche of Ancona. However, rather than the two already accepted chartmakers from that family, Conte Ottomanno Freducci (supposedly active 1497-1539) or his son, Angelo (fl. 1547-56), she proposed instead a radical revision of the identity and sequence of Freducci chartmakers. An earlier Count Lodovico (recorded up to 1478 - Licini, 2005, pp.117-19) was her choice as the author of the Jesi chart and she asserts that the person who signed the 1497 chart in Wolfenbüttel as 'Loctomanni' was actually Oliverotto (1472?-1502). She considers 'Ottomanno' to be a corruption of 'de lo etomanno', the region over which the Freducci family had jurisdiction (2007(b), pp.264-). If that were confirmed it would of course mean that all the charts dated after 1502, when Oliverotto was murdered on the orders of Cesare Borgia (Licini, 2007, p.30), would be the work of one or more chartmakers intermediate between Oliverotto and Angelo - and hence a significant addition to our knowledge of Anconitan chartmaking.

There is no published version of the Jesi chart in high resolution and the poor state of preservation resulting from its secondary use in a binding makes it difficult to see small details. It is also unfortunate that among the discarded sections are those in the Atlantic that are most useful for diagnostic purposes. Nevertheless, the outlines of the underlying chart do look very similar, in general terms, to those of Benincasa, which were, in turn, to be copied by members of the Freducci family for a century. Although the chart's suggested date of 1464 would fit in well with Benincasa's career, minor deviations on the Jesi chart from his remarkable stylistic consistency rule out his authorship. The lack of the habitual Benincasan scale bar in the surviving top margin is also significant.

If the Jesi chart had a signature and date it would been on the missing section; it has no meaningful provenance; and it cannot, stylistically, be confidently assigned to any known chartmaker. That is the very uncertain background against which Licini weaves a highly complex historical story with what seem, at times, to be tenuous interconnections. In what might be considered a circular argument, the chart's dating (1464) is seen to depend on a direct connection with Pius II and his failed crusade from Ancona, although that is not, in my view, adequately explained. The supposed stylistic similarities with the work of Ottomanno Freducci (which would normally be considered to refer to the work of a single chartmaker, thought to have been active between 1497 and 1539) is here used to introduce at least one earlier member of a supposed Freducci chartmaking dynasty, one who died on or after 1478. It is, however, highly unlikely that Benincasa and one of the Freduccis would have been working in the same style, and so close to the date when Benincasa is known to have started his chartmaking career (1460).

A further argument is adduced for there being a Freducci active as a chartmaker in the 1460s, on the basis of the chart in Weimar (now Klassik Stiftung Weimar). Today, it is possible to read, imperfectly, only ' Comes...' [or Contes] at the start of the authorship inscription but a century or more ago much of the same wording as that found on the clearly signed Ottomanno Freducci chart of 1497 in Wolfenbüttel could apparently be made out. There is little reason to doubt that both are by the same person but the Weimar chart's date is another matter. 19th-century scholars read it as MCCCCXXIV but W. Ruge (1904) saw instead it starting MCCCCLX ... '1424' was clearly wrong, but attempts to turn Ruge's '1460+' into c.1460 or just 1460, should be discounted. The 1497 chart's date was written as MCCCCLXXXXVII, which means that the Weimar chart, presumably repeating that formula, could have been drawn as late as 1499. There may be good reasons for thinking that the Weimar chart might be earlier than that in Wolfenbüttel but I am not aware of those. There is certainly no justification for assigning the Weimar date to the 1460s, and hence for seeing it as contemporary with the date of c.1464 that Licini suggests for the Jesi chart. Nor, in my view, is there compelling evidence anyway for the fragment's Freducci authorship.

I did notice what appears to be one small, later addition to the Benincasan norm, at the mid-point of the Black Sea's east coast. Marking the outlet of the Enguri river, beside the red-named faxio, there is an island, here as usual coloured red, shaped like an old-fashioned spanner or a catapault. In Benincasa's work up to 1469, the chronological point at which the Pujades DVD ends, the island is always left unnamed, as it had been by other chartmakers. On the Jesi chart, however, it is labelled, in the sea, as isola de colchi. I am grateful to Anton Gordyeyev for drawing my attention to other instances of this name, particularly on the 1474 Benincasa atlas, where it is included in the inland sequence. That is not repeated, however, on the Andrea Benincasa atlas of 1476 in Geneva. Most other instances of Colchi in the later 15th and 16th centuries clearly refer to the province of that name. The name does not appear on the 1497 Freducci chart in Wolfenbüttel.

One further idiosyncracy, to which Professor Licini has drawn my attention, is the treatment of Fiume [i.e. River] Esino, close to Ancona. Amost all early charts ran the elements together into a single word, for example, fiumexino. The Jesi's author presents it as f. [?]xiro. This form is not recorded for any chart up to 1469, has not been seen on Benincasa's work, and does not appear on samples of the Freducci output I was able to consult.

The handwriting of the Jesi chart is difficult to discern from the published illustrations. The best that I have seen is the enlargement of 'Armenia Minor', showing part of the northern coast of Turkey (Licini, 2000, p.60) but there are also useful details in her 2008 chapter. The hand is neat and without any obvious distinguishing characteristics. Further analysis, with a range of comparative scans of adequate quality, would be needed to test the suggested Freducci attribution.

[I am very grateful for the generous assistance provided by Professor Licini even if I remain to be convinced by her arguments.]

Literature:
See, separately, under Bonasera and Licini in the Bibliography
Giovannozzi, D. 'Freducci, Conte', in the Dizionario Biografico degli Italiani [1982 or later - accessed 22 July 2011].

Lucca, Archivio di Stato, Fragmenta Codicum, Sala 40, Cornice 194/1, A chart fragment, 60x30 cm, recovered from a binding.

The chart was discovered in 2000 in the binding of a book of notarial records (1611) belonging to Giovan Paolo Giampaoli, who was active in Lucca 1595-1611. It has been trimmed at left, top and right. The surviving coverage is of the entire Mediterranean (except the Holy Land), part of the Black Sea, and the Atlantic coasts between southern England and Morocco. Among its distinctive visual features are the 26 coastal towns 'represented by exaggeratedly large pictorial signs and flags', and, unprecedented on a portolan chart, fragments of scales and winds with human form.

Philipp Billion, whose 2011 article in Imago Mundi is the first study to be devoted to this chart, suggests a late 13th or early 14th date, or more specifically, 'before 1327'. He argues that its place of production might have been Gaeta or Pisa. The elements cited in favour of an early date are the coastlines of southern England and the Bay of Biscay, which are seen to fall, in developmental terms (which is not necessarily the same thing as a chronological sequence), between the representations found on the Carte Pisane and Vesconte's earliest treatment of those regions in 1313. However, it is the case that the outline given to England's south coast and its offshore islands is markedly different from those to which it is compared. The shape of Italy on the Lucca chart is shown to match better with the Carte Pisane than with the Cortona chart or Vesconte's first chart of 1311.

In the Imago Mundi article the toponymy was compared with the Pujades listings for the coasts of Valencia and the eastern Adriatic (though not his listing for the western coast) but, besides finding a number of atypical features on the Lucca chart, the results from a relatively small sample (108 names) were inconclusive. A few names linked the Lucca chart with the Carte Pisane, among them oriola (which Pujades saw as synonymous with the Segura river), a toponym that was not otherwise noted on any pre-1470 chart. Some names are otherwise first recorded on Vesconte's earliest atlas of 1313 or, in one case, on a chart of c.1325. Another name is reliably dated first to the 1339 Dulceti chart (Mazzorbo), while Cadaqués first appears on a dated chart in 1403. There is a need for a full analysis of the chart's toponymy.

City signs are first seen on the Cortona chart and, from about 1320, Vesconte's atlases, to which flags were also added. However, the Lucca's are visually very different.

Although the surviving sections of the chart cover only parts of the usual portolan chart region, the Colour & Shape Analysis was able to extract some useful information. In the first place, just two colours were used for the islands, red and blue. After the two supposedly earliest surviving portolan works, the Carte Pisane and the Cortona chart, both uncoloured, the earliest experiments with island colouring involved no more than two tints (on the undated Riccardiana chart in Florence) and three on the 1311 Vesconte. Billion identifies a number of other features that point to an early stage of development, at whatever date the chart was actually drawn. Besides the Lucca chart's restricted colour range, the C&SA findings provide other support for an early model in its island shapes.

Backing up the interpretation of the Lucca chart as containing features associated with the earliest phase of (surviving) portolan chart development are the simplified geometric shapes of some of the islands considered in the C&SA. In most cases the Lucca chart offers a broadly rectangular outline with a few indentations, of a type otherwise found only on the Carte Pisane (e.g. Zakynthos) or on the Carte Pisane and the early Genoese chart in the Riccardiana Library (e.g. Skyros). For Majorca, Corfu, Ithaca, Lefkada and Limnos, the Lucca's formulaic outlines are unlike any seen elsewhere, just as its Nile delta has a single right-hand island represented uniquely as a simple rectangle. Keffalonia, on the other hand, is already given a broadly recognisable, certainly non-conventional, shape on the Carte Pisane and the Lucca chart, in clear distinction to the Cortona chart's simple, dog-bone outline.

Insofar as these could be made out, there was no sign on the Lucca chart of any C&SA conventions introduced after the period of its suggested construction. For more on the earliest charts see Anonymous works and the question of their attribution to individual chartmakers or to their supposed workshops. Unfortunately, no comparison was made by Billion with the earliest surviving Genoese chart, that in the Riccardiana Library, Florence, which is certainly from the early part of the 14th century and could even pre-date Vesconte.

Literature:
The following article should be seen as a valuable, preliminary study of a highly unusual chart, which deserves fuller analysis in the future.
Philipp Billion, 'A newly discovered fragment from the Lucca Archives, Italy, Imago Mundi: the International Journal for the History of Cartography 63: 1 (January 2011): 1-21.

’The Lucca Chart, anonymous Italian, around 1327’ (a note by Sima Krtalic for the MEDEA ‘Chart of the Week’, 21 January 2021, referring to the Medea-Chart record).

A detailed reassessment of the Carte Pisane: a late and inferior copy, or the lone survivor from the portolan charts' formative period? (March 2015) and Brief notes on the main documents discussed.

Modena, Biblioteca Estense, C.G.A.1 - 'Catalan (Estense) world map'. Diameter 113 cm.

This very well known and much-studied map/chart was not listed in my Census nor considered by Pujades. I now realise that it should have been included.

A detailed separate note in Anonymous works and the question of their attribution to individual chartmakers or to their supposed workshops concludes that it was, as previously suggested, drawn by Roselli, that at its heart it is a portolan chart, and that it can be dated to c.1462-4.

For an enlarged scan see the MEDEA-Chart database.

E.24

Oxford [now London], Daniel Crouch Rare Books (2011). [Update: on 5 September 2014 it was announced that the fragment had been acquired by the City of Barcelona.]
[A version of the following extended text appeared in Daniel Crouch Rare Books LLP, Oxford, Catalogue II (September (?), 2011), item 1. ISBN: 978-0-9567421-1-7.

A fragment (21x31 cm) of a Catalan chart, in the same hand as that of the two signed by Guillem Soler, and evidently slightly earlier than those, covering the lower half of the Iberian peninsula and the north-west coast of Africa. pre-1385?

Physical description
The surviving fragment measures 209 x 312 mm at its maximum dimensions. It represents a vertical strip taken out of the lower left quarter of a dismembered chart, and covers half that chart's height. The coastlines shown include the lower half of the Iberian peninsula and the north African coast, almost from the usual termination point in Western Sahara, and then as far east as Oran in Algeria. The size of the original chart can be estimated at about 64 x 105 cm. The fragment survived because it was re-used in the binding of an octavo volume measuring 164 x 108 x 65 mm [see 'The host volume' below]. It is a sobering thought that there could be at least another seven such sections from that original chart, perhaps used for the bindings of related works. Maybe one or more might appear in future.

The full chart would have had two rhumb line networks, with the centres of the hidden circles placed, respectively, in northern Spain (just off the top of the fragment) and the Aegean. The twin centres would have met north of Sicily. The bottom of the fragment may well represent the original lower edge. On the analogy of the 1385 Soler chart discussed below, the line running beneath the scale bar would represent the lower margin of the chart's central section. At the west the chart would have continued further to the south so as to include the coast down to Cape Bojador and the Canary Islands.

It is argued below that this chart would probably have included the signature of the Catalan chartmaker Guillem Soler (fl. 1368-1402), who worked in Palma, Majorca. He is known from two, visually very different, signed productions, representative of the range of Catalan work: one a plain chart dated 1385 preserved in Florence (Archivio di Stato, C.N.3 - Pujades C 17), the other an ornate version with an undated inscription in the Bibliothèque nationale de France, Paris (Rés. Ge. B 1131 - Pujades C 14). The fragment must come from a work of similar size to that of the 1385 chart. The double-page illustration in Pujades (2007 pp.158-9) gives a very good idea of what the fragment would have looked like in its original context.

Significant features
The fragment has four distinctive features which help confirm the Soler authorship: sadra, the 'Plages arenosses...' inscription, the scale bar, and the single town vignette.

sadra
At the bottom of the chart, set into the unknown interior of Africa, is one large name, written exactly thus sadra:- It appears in precisely the same way on the 1385 Soler chart but the relevant area is missing from the undated one. sadra was not seen on any other 14th-century chart, although, being close to the edge of the vellum, this area is sometimes missing.

'Plages arenosses...' inscription
Nearby is a truncated inscription. This appears in full on the 1385 Soler chart but only its right-hand section is preserved on the undated Soler chart. The folding table in Ramon Pujades's 2009 study of the 1439 Vallseca chart transcribes 29 legends from 11 Catalan charts between 1330 and 1439. Several start 'Plages arenoses... [Sandy beaches]' and a few have a similar wording. However, the Soler inscription is unique, and the fragment follows that form exactly. The missing sections of text are supplied here in square brackets from the 1385 chart; the contractions are filled out in italics:

["Plages arenosses desertes si]no de peschados los quals dien sisotz X milles en mar [trobaretz X passes de fons per] tota esta costera segons que seretz en mar mes homenys"

Two small sections - involving milles and per tota esta coster - could not be clearly read on the 1385 chart but are here confirmed. The eight surviving words in the equivalent inscription on the undated chart are the same as the other two, underlining that this wording is specific to Soler and was carefully repeated by him.

In June 2011, it was possible to examine the chart in person, albeit through glass. It now seems that there is a second inscription running in from the West African coast, between the red names Gutzolla (just above the horizontal rhumb line with the sequence of very visible holes) and Mensa. The unusually long name between them is apparently porto meseguinam - though only the porto and the final three letters are legible. To the right of that is what looks like a black 'a' and, along and below, what appear to be faint traces of lettering. At exactly that same point on the undated Soler chart in Paris there is a six-line inscription. This is transcribed by Ramon Pujades in his 2009 work, beginning 'Aquest pas es apelat val de dara ...'. It seems highly likely that a version of that same inscription was originally placed here. {This paragraph added 22 June 2011}

Scale bar
The scale bar, with its double line above filled in with light yellow-brown wash, is typical of Catalan work of the second half of the 14th century [see Pujades, 2007 p.220 for a composite display]. What is not found other than on Soler's work is the way that the long scale strip running across the middle of the bottom of the chart is bounded by north-south rhumb lines to create a block of empty space beneath. This represents the central portion of the chart's lower border. We can therefore assume that the fragment's truncated scale bar would have run across to the equivalent position at the right side, and that it would have been repeated at the top.

The fragment is unusually devoid of decoration. It does not even contain the name for the south-west wind Libeccio within the usual circular frame. Even if the fragment is a visual disappointment, its lack of non-functional ornament makes it far more likely that this was a very rare survivor of the type of unadorned chart designed for use at sea. As such, it would have been priced down for the large seafaring market. Far more than the highly ornamented landsmen's productions, which survive in disproportionate numbers, this is faithful to the practical purpose of a portolan chart. The long water stain which runs across the chart (roughly NW-SE) must have been present on the vellum before it was cut up (and folded inwards) to form the binding. It is thus valuable evidence of the physical state of a chart at the time it was abandoned from over-use. {The previous two sentences added 22 June 2011}

Town vignette
The fragment's sole illustrative feature, the stylised vignette of trimssi, denotes Tlemcen in northwestern Algeria. This was a major trading centre in the Middle Ages, both for the cross-Saharan traffic and for that along the Maghreb littoral. Soler's almost equally plain chart of 1385 added equivalent vignettes for Marrakech (marochs) and Granada, which could have appeared here. Had a more elaborate model been followed, such as Soler's other, undated chart, the fragment would have had several flags, the Atlas mountain range and river courses as well.

Soler's formula for Tlemcen - since his three versions are indistinguishable - was to place a pair of towers either side of a central building with a tall minaret, all within a walled enclosure. This was viewed from a slight elevation, giving a sense of perspective and allowing the back of the inner wall to be picked out in red. An equivalent view had appeared on the earliest Catalan charts but neither that, nor the contemporary productions of the Cresques atelier, matches Soler's style. Despite being a luxury item, the Catalan Atlas of c.1375 pared the image down to a single central minaret. This was placed, as expected, within the walls but for some other cities the central feature was placed outside the town and behind it. Three of the four charts assigned to the Cresques atelier repeated that implausible device, while the fourth has a different design altogether (see Pujades, 2007, p. 63), as does later Catalan work. [I am grateful here for comments from Sandra Sáenz-López Pérez.]

Toponymy
The fragment includes the coasts of southern Iberia, from Porto in Portugal to Valencia in Spain. For north-west Africa, it runs from [allue]t nul, five names short of the usual terminal point, Cape Bojador (buyetder), in modern Western Sahara, up to Seuta and then east as far as tigis[mach], just beyond Oran in Algeria.

It is unfortunate that the area covered by the fragment barely overlaps with the two regions whose toponymy has been studied in detail by Pujades, i.e. the coasts of Catalonia & Valencia and the Adriatic. Only twelve names south of Valencia itself can therefore be checked against his composite listing. For the remainder, the names on the fragment have been assessed against the ongoing composite listing of Significant Names (a Microsoft Office Word 2003 table).

Any dating conclusions derived from place-names alone need to be prefaced with a note of caution. First, on account of the small size of the sample, second, because neither Iberia nor North Africa were areas of great toponymic development in this period, and, third, since Catalan chartmakers were slow to adopt the Italian names introduced in the early 15th century. That said, some interesting conclusions can be drawn.

Like the two signed Guillem Soler works, this fragment contains no place-names so far noted as having been introduced from 1367 onwards. The only exception relates to the Catalan Atlas of c.1375 (on which see more below). The twelve names south from Valencia to Guardamar are sufficient (with the supporting evidence given above) to show that this is not the work of Soler's contemporaries and neighbours in Palma, Majorca, i.e. those thought to have been associated in a joint workshop with Cresques Abraham, the supposed author of the Catalan Atlas.

Soler's alteya form (repeated here) is otillia on the Cresques atelier charts, and his cantera is given by them as alacant. Though the name is only partially legible here it certainly starts with C not an A. Most significant is the omission on this fragment and the two Soler charts of the flum de segura, found on almost all Catalan charts between 1339 and the second half of the 15th century. Among the few other charts to omit this are the two produced by Soler's successor Rafel Soler. The omission of the Riu Segura is thus one of the defining characteristics of productions of the Soler family. [For a general comment about the distinctive toponymy of the Cresques group of converts from Judaism compared to that of the Christian Soler family see Pujades, 2007 p.492b and follow the reference at the end of note 87.]

With the attribution to Guillem Soler established, how can we relate the fragment to the two signed charts already known: the one in Florence reliably dated 1385 and the other, in Paris, assumed to be earlier, conceivably as early as 1368? What place-name differences can be observed between the three works?

Three of the toponymic forms suggest that the fragment could be the earliest of them all. Cullera, next to Valencia, was conveyed as cugera on Catalan charts from 1330, until, from the late 14th century and on through the 15th, it changed to cuyera or, more usually, culera (Pujades, 2007, pp.394-7). The two signed Soler charts use the later form; this fragment has cugera . Another name, Riffiene, conveyed by rif, next to Seuta at Africa's north-west tip, seems to have appeared first on the Catalan Atlas. It is included on both the signed Soler works but not this. Another Catalan Atlas innovation, vacar (between Cadiz and Tarifa) [as distinct from torre de vacar (just east of Algeziras)] is certainly absent from the fragment but the two signed works are not clear enough to read. choria, a little to the west of Cadiz, is the only one of three relevant names first seen on the Catalan Atlas definitely to appear on both the two signed Soler works and on this partial chart.

The final, and most significant indication comes from the name to the east of the Algerian town Honaine (one). Generally, from early in the 14th century onwards, the toponym that appeared at that point was gordanea. It seemed to have been Soler who introduced a relatively short-lived alternative, muguron, identified by Pujades (2009, p.155, no.69) as Ile Mokrane. Some variant of that - in neither case is the reading clear - was included on the two signed Soler charts, but the fragment displays the earlier go[z]da[---] form. That muguron formed the standard for Soler in his later career is evident from its perpetuation in the work of Rafel (certainly on the signed work in Berlin and apparently on the attributed Paris chart). This is the strongest evidence pointing to the fragment being earlier than either of the Guillem Soler charts already known.

Handwriting
The fragment is evidently 'by' Soler, in the sense of the authorship of its style and content. But might it have been the work of one of the other four members the Soler chartmaking dynasty started by Guillem (Pujades, 2007, p.492; 2009, pp. 312-15) - for a convenient summary see Here (look under 'Soler')? Besides Guillem, all the others seem to have been active in the 15th century. Only one, Rafel Soler, has left surviving productions and neither of his two charts has a readable date. Despite the continuation of some toponymic conventions from Guillem, Rafel's own style is noticeably different. There is therefore no other Soler family member to whom the fragment could be attributed.

The fragment's handwriting confirms those findings. Dr Ramon J. Pujades i Battaler has examined a scan of this fragment and has made the following statement: "This fragment was copied by the same hand that wrote the nameplaces and legends on the two charts signed by Guillem Soler".

Dating
What can be said about the fragment's likely dating? One of Soler's two charts is clearly dated 1385, the other never had a date. That Pujades gave it such a large possible window, 1368-85, reflects, on the one hand the earliest evidence of Guillem's activity in 1368 (Pujades, 2007, p.491b) and, on the other, the date of one of his charts. But in what ways can the two charts be chronologically distinguished and why might the complete undated chart not be later than 1385, given that Guillem could theoretically have continued working to the end of century?

The dated chart is plain, like the fragment; the undated one is visually quite different, being ornate and using gold leaf. But the toponymy of the two signed charts, as set out by Pujades for the Adriatic and Valencia (2007, pp. 374-85, 394-5), shows no significant differences. Why could both not be close to the same date, i.e. 1385?

The fragment's few toponymic variations (discussed above) lead logically to the conclusion that it is earlier than either of the others. On the understanding that the undated Paris chart could be re-dated to c.1385, the suggested date for this partial survivor might be c.1380 or even a little earlier.

The host volume
On page 10 of the Daniel Crouch catalogue there is an excellent photograph of the verso of the vellum, distinguishing clearly the discoloured central section, which formed part of the outer binding, from the lighter strips that would have been folded in behind. Across the middle, i.e. along the spine of the volume, there is a Latin inscription. I am grateful to Ramon Pujades for the following explanation of what that is likely to indicate.

"It reads Flores Teologicarum Questionum, which seems to correspond to the work known as Flores theologicarum quaestionum, in quartum librum sententiarum. This was a relatively common theological book written by Josep Anglès [or Josephus Angulo in Latin] (1550-88). He was a friar professor at Valencia University in the second half of the 16th century, and was created bishop of Bosa (Sardinia) in 1586."

The book was first published in Cagliari, Sardinia in 1575, with subsequent editions in (at least) Antwerp, Lyon, Madrid, Rome, Turin and Venice up to (at least) 1616. Pujades is cautious about a definite identification of the printed work that was enclosed by this chart fragment for perhaps 400 years. However, neither Google searches nor WorldCat throw up other instances of the first three words of the title. The identification sounds highly likely, particularly as the size given in one instance for an edition of the octavo Anglès volume matches almost exactly the height that can be deduced from the fragment's verso.

Since there is no sign that this fragment was itself cut down from a larger, previously used section of the original chart [note the lack of discolouration on the side panels], and its current use could not date from before 1575, some two centuries would seem to have intervened before the Soler chart was dismembered. {This 'host volume' section added 28 October 2011}

Provenance:
This is not known

Literature:
Pujades (2015, 2009, 2007)

Sotheby's 11 June 2007, 'Russian Books, Maps and Photographs, Lot 201 [withdrawn from the sale].

The catalogue note reads:
"Untitled manuscript portolan chart on vellum centred on the Caspian Sea. [Venice, late 15th century], 390 x 524 mm. (at greatest extent), ink and colour on vellum, folding map (originally binder's waste), some wear and discoloration at folds (with loss).

This interesting chart is orientated with South to the top of the map and covers: south of the Black Sea and Caspian Sea (to Baghdad or Tehran/Persepolis), west as far as the Crimea, east to the lands beyond the Caspian (but not as far as the Himalayas) and north to Kharkov.

In the style of the Catalan charts are 33 large-profile illustrations of towns and cities, a portrait of "lo gran amperador de tartarja", the Asktrakan delta is shown in red, green and gold and the rivers Don, Volga and Tigris and Euphrates are clearly marked with the towns of Volgograd and Saratov shown."

The catalogue includes a half-page colour illustration, which is not available via the Internet.

The size of the illustration does not allow the place-names to be read or the finer details to be made out. The following observations are therefore tentative and incomplete.

The Sotheby's cataloguer concluded that this was Venetian work of the late 15th century. The reasoning behind those two judgments is not clear. Unless Venetian elements were detected in the handwriting, or the orthography of the toponymy reveals dialect forms that point to that city, it is more helpful to assume that this is a Catalan production, since the illustrative elements are typical of Majorcan work. The transcribed label, 'lo gran amperador de tartarja', is in Catalan. There seem also to be grounds for suggesting that the date might be a little earlier.

Extent of the chart
The fragment, clearly removed from a binding which has left the marks of folds and sewing holes, represents the top right-hand quarter (or perhaps one sixth) of the original. It appears to be oriented to the south, as indicated by what can be seen here, but it is likely that the illustrations and their accompanying legends would have faced the side of the chart closest to them, hence giving a multi-orientation to different parts of the chart. The Catalan Atlas of 1375 is a good example of this convention, which was followed in subsequent Catalan work and on the Genoese chart of Battista Beccari (1426). The Sotheby's illustration follows the logic of the fragment in placing south at the top.

However, for convenience the chart will be described here in terms of a northward orientation. The outer rule marking the top and right-hand edges of the original can be clearly seen, with a simple bar scale set into the top margin. At the bottom, the fragment was cut fairly precisely along the line that ran east-west through the centre of the rhumb line network. In other words, half the height of the original survives. But what is unusual is that the chart extended to the east beyond the Black Sea, which has prompted the insertion of a second incomplete rhumb network to the right. The area of join, between the Black and Caspian seas, is mathematically complex and unusual. The normal practice was just to extend the existing rhumb lines beyond the 'hidden circle'.

Even on charts that include the Black Sea the eastern edge of the rhumb line network normally falls near Palestine. In this case it is some distance to the east of that. If the original from which this fragment was cut had twin rhumb networks there could have been two equivalent-sized sections to the west. In that case, there would have been room for the full normal extent out into the Atlantic. Alternatively, this could have been one of those (not unusual) charts that stopped at west Spain, in which case a single network might have sufficed.

Given that the surviving fragment covers half the original height and measures 39 cm, we can be fairly sure that the original vertical dimension was close to 78 cm. This is larger than average, though not excessively so, when compared with the sizes provided by Pujades for all surviving charts pre-1469 (2007 pp. 63-5). By analogy, a height of 78 cm might suggest a width of around 115 cm. The Caspian extension could imply a greater lateral dimension but the normal ratio of height to width will usually reflect the standard shape of the animal skin from which the vellum was taken.

Toponymy
It is only the fragment's western quarter that includes coastal names. The vertical cut at the left-hand side runs down the western side of the Crimea and across Asia Minor, retaining just the eastern halves of Turkey's north and south coasts. While toponymy, and particularly the introduction of new names, can be a useful pointer to date, the Black Sea was one of the most static areas. Between the Catalan Atlas in 1375 and 1500, no more than five fresh names were introduced by Catalan chartmakers around the Black Sea. For the south coast of Asia Minor there was a single name. So, even if the toponymy could be read, the mere incidence of names is unlikely to be revealing. It should, though, be able to confirm the chart's place of origin.

Illustrative details
The Sotheby's cataloguer noted that there are 33 city vignettes. This is a standard feature of Catalan charts, and of the occasional Italian ones that imitate their style. However, the profiles here are more varied and lifelike than the formulaic Catalan work from the late 14th and early 15th century: for example, the charts produced by the Cresques atelier and the two members of the Viladesters family. In this case they are carefully drawn, mostly individualised and, in a few cases, more elaborate than usual. What is noticeable is the narrowness of the towers, enabling far more to be portrayed. The thin vertical blue line down the side of the towers is a device intended to indicate shadow, first seen on the work of Vallseca.

The style of the towns is generally reminiscent of the later Vallseca charts (1447-49), though more sophisticated. The closest comparison, perhaps, comes with the chart drawn in 1426 by Battista Beccari, whose father Francesco had been the illuminator who worked with the Cresques on the large wall-chart commission in 1399. Given that the illustrative work was presumably carried out by specialists who might have been employed irregularly, and because Catalan practitioners continued with broadly similar illustrations into the second half of the 16th century (see, for example, the 1563 work of Matteo Prunes, illustrated in Rosselló, 1995, p.99), it may be unwise to try to identify the work of a chartmaker from the town vignettes alone.

The catalogue also mentions the larger of two ruler portraits to the north and east of the Black Sea, transcribing the Catalan legend of the larger one as 'lo gran amperador de tartarja'. This inscription is read more plausibly by Ramon Pujades as 'lo gran cha amperador de tartarja', the form found on the 1482 chart by Jaime Bertran. Such rulers were usually shown unframed, as is the case with the other, scimitar-wielding figure to the west, but the Tartar ruler is shown as if painted onto a rectangular hanging canvas. The 1439 Vallseca chart provides a good example of that same device.

Above the Sea of Azov it is possible to make out a third, very faint picture. The Florence chart attributed to Vallseca (Bib. Naz. Centrale, Port. 16 - see further below) labels this twice, on repeated illustrations, as 'alans'. In her 2007 thesis (pp.653-5, Figs 93, 172-3), Sandra Sáenz-López Pérez described the characteristics of the Alans, naked savages who used hunting dogs to attack mythical beasts or lions. It would seem that the various representations of the Alans, found first on the 1413 Viladesters chart, then on the Florence chart (Bib. Naz. Centrale, Port. 16) and the Catalan Estense world map by Roselli, may feature a lion being grabbed by a long-haired man. There is perhaps a reference to Hercules here.

The fragment is adorned with around 21 flags, though they are not clear enough to distinguish. Nor are flags, generally, a reliable dating guide. There is no obvious match between the scale bar and any of those illustrated by Pujades (2007, pp. 220-1).

Reading the faint outline of the Caspian Sea is not easy. On other Catalan work that includes the sea it has been rendered highly visible by means of parallel wavy blue lines. This might possibly suggest that this chart was left uncompleted, though the illustrations would have presumably been the last elements added and it seems that there is already a full complement of those.

Caspian Sea
The fragment's most surprising feature is its inclusion of the Caspian Sea. Because the chart's normal range was carefully extended to the east to accommodate that and, it appears, little else, this was clearly thought important to its creator (or commissioner). Few portolan charts extended beyond the Black Sea and when they did they rarely showed much detail.

A world chart is of course obliged to show the Caspian, hence the Catalan Atlas (at its heart a portolan chart) is the first to include it. Ptolemaic maps had shown the Caspian as an inland sea, presented broadly as an oval 'landscape' shape. By contrast, most medieval maps treated it as a gulf of a northern ocean. The Catalan Atlas depiction, broadly followed here, is landscape rather than portrait in presentation, and slightly larger than the Black Sea but not wholly dissimilar in shape. This form is described by Cyrus Alai as superior to all Ptolemaic and European maps produced before the mid-17th century (General maps of Persia 1477-1925 (Leiden: Brill, 2005) p.3).

The Caspian on this fragment, although simply drawn, is quite complex. Its outlines are a close match to those on the Catalan Estense world map/chart, confidently attributed to Roselli and dated by myself to 1462-4 (see Here). There appear to be no names around its shoreline besides, it seems, identifying labels placed next to three town vignettes. Where a large river is shown entering the sea at the north-west (presumably the Volga) the complexity of the delta is conveyed by a three-coloured checkerboard pattern. This can be seen also on the Catalan Atlas, was repeated on the 1413 Viladesters chart and, simplified, on the Catalan Estense world chart. This analysis should have mentioned the 1455 Bartolomeo Pareto chart (Pujades C 57), which includes the western half of the Caspian in a form that closely imitates the Catalan Atlas {This sentence added 18 July 2018}

A fragmentary version of Ptolemy's Geographia in the British Library (Harley MS 3686) was described by Marica Milanesi (1996), as "one of the earliest examples of the synthesis of portolan chart, Ptolemaic map and medieval mappamundi...". The eastern section of the Caspian Sea (Fig.7) is included, with an outline that looks broadly similar to that on the Sotheby's fragment. The author suggests a Venetian provenance and a date in the period 1436-50.

Given the close proximity of the Black and Caspian Seas on this fragment, the 1439 Vallseca chart would have had room to show at least part of the Caspian, but chose not to. The only other early Catalan chart to include the Caspian is an anonymous work in the Biblioteca Nazionale Centrale, Florence, Port. 16 (C 41 on the 2007 Pujades DVD), attributed by Pujades to the Vallseca workshop around 1440 and dated by Sáenz-López Pérez to the period 1439-60. However, its Caspian is arranged vertically and given a noticeably different outline. Quite distinct again are the two depictions in the unsigned Medici Atlas, considered by Pujades to be a Genoese work of the second quarter of the 15th century. The world chart (Sheet 2) has a relatively simple outline oriented northwest/southeast, while the right half of sheet 3, devoted to the Caspian Sea, contains a more elaborate outline with its own considerable toponymy. However, there is clearly no connection between that and the various Catalan forms.

Authorship and dating
The fragment's debt to the 1375 Catalan Atlas for its Caspian Sea, as well as the way that other elements find echoes in, for example, the 1413 Viladesters chart, the work of Vallseca (1439-49), and that of Roselli in the 1460s, points to Palma, Majorca as its almost certain place of production. Given that Catalan conservatism means that some conventions introduced in the late 14th century were still present in the 16th, dating is more difficult but somewhere in the second half of the 15th century seems a reasonable guess. The unusual sophistication of the artist responsible for the town symbols may indicate somebody whose work on portolan charts has not otherwise survived. There does not seem to be justification for assigning the fragment to any of the Catalan chartmakers who are already known.

However, as stated at the outset, this analysis is necessarily provisional until it becomes possible to examine the original or a high quality reproduction of it. If that happened - and there is no news of the fragment's current whereabouts - other expertise, particularly relating to its iconography, could be brought to bear on this most unusual discovery.

I am grateful for suggestions from Ramon Pujades and Sandra Sáenz-López Pérez, but they were only able to see poor copies of the relatively small illustration in the Sotheby's catalogue. I may occasionally have diverged from their opinions.

Provenance
This is not known

{This entry added 21 October 2013}
National Library of Egypt, Cairo, Map Library, Portulans POR 1

The 2009 description of this chart (82 x 125 cm) provides a date of 1312. The reason seems to be because it was thought that the Carte Pisane in the Bibliothèque nationale de France was 'très voisin de cet exemplaire' [very similar to this map], a comment apparently inspired by George Kish, La carte, image des civilisations (1980). The Cairo chart is described as 'ancienne carte nautique de la collection, d'origine pisane' but the ornamentation referred to - figures of humans and animales, wind roses, etc - are noticeably dissimilar from the supposed model.

There is an accompanying scan which can be enlarged (clicking on 'Image gallery' or the thumbnail produces the same result). Unfortunately it immediately moves out of focus and little detail can be discerned. What can be made out, e.g. a saint's (?) figure in the neck and the scale set out on a partly furled ribbon, suggest a 16th-century production. The scale might enable the chartmaker to be identified. The Atlantic is excluded but the Black Sea is shown complete.

This '1312' chart could probably have been ignored were it not for the fact that this misinformation is readily available via the internet on e-corpus, 'a collective digital library'.

Vernadsky National Library of Ukraine (Kiev), Manuscripts Department [It is evidently not included in the current online library catalogue and its manuscript number is not known, unless the earlier notation, noted below, is still used.]

Grazioso Benincasa. 6-sheet atlas, 33.5 x 42.0 cm, glued to boards, with paired calendars. Signed on the central Mediterranean sheet: GRatiosus Benincasa Anconitanus / Composuit Veneciis MCCCCLXXIIII [1474]

In May 2015 I was alerted by Anton Gordyeyev (Gordeev) to the existence of a second portolan atlas by Grazioso Benincasa bearing the date 1474. It is definitely different from the 1474 version already known (National Széchényi Library, Budapest (Manuscript Collection, Fol. Ital. 8. - formerly Magyar Memzeti Muzeum, Budapest). That had come from the collection of Count Ferenc Széchényi, who founded the Hungarian National Library in 1802. The authorship inscription in the Kiev atlas omits the words 'Anno Domini' before the date, which appear in the Budapest volume. For much of what follows I am indebted to Anton Gordyeyev, and to sight of his forthcoming English-language article for the Vernadsky Library magazine. [For the Ukrainian text see Here.

The atlas rediscovered in the Vernadsky National Library of Ukraine in Kiev was described in a published volume that seems to have been wholly overlooked afterwards, at least by those interested in portolan charts, and certainly outside the Russian-speaking world, where it was considered to have been lost after 1917: Veniamin Aleksandrovich Kordt, [Materialy po istorii russkoy kartographii]. Series 1, Volume 1 (Kiev, 1899), p.15.

From two printed ex libris (p.360) stuck-down inside the cover, it can be learnt that the atlas had earlier been in the collection of Count Michal Jerzy Wandalin Mniszech (1742-1806) - 'Ex Libris Mich: Comitis Vandalini Mniszech'. Anton Gordyeyev has been able to date this engraving no earlier than 1780 because it includes the Order of the Apostle St Andrew, the third of the orders of St Stanislaus which Mniszech was awarded. Pasted over that is the later ex libris sticker, stating in Russian, 'Manuscript Division of the Imperial University Library' [of St Vladimir], No.166. A label on the atlas spine today carries the numbers VIII. 189M (horizontal line) 166.

Anton Gordyeyev has kindly supplied me with a copy of the sheet bearing the paired Black Sea and eastern Mediterranean charts reproduced in monochrome in that 1899 volume. The illustration was placed alongside a very brief description in Russian, which adds no useful information. Paper damage at the north-west corner of the Budapest version of the Black Sea, absent from the Kiev example, provides further confirmation that two different works are involved.

This is a major (re)discovery. It is the second portolan atlas to be added since the 1986 census, and one of only two signed portolan works in this supplementary listing.

The atlas comprises a calendar and six charts, in the usual arrangement:

A pair of calendars occupying the two halves of a folio, whose first table runs from 1470 to 1489, and the second covers the period 1470-1569 (in both cases apparently like the Budapest atlas). [Interestingly, a later hand has provided comments below, in Italian, while updating the first calendar for the period 1698-1716.]
1. Eastern Mediterranean and Black Sea
2. Central Mediterranean (including the Aegean) [with the authorship inscription]
3. West Mediterranean
4. North-west Africa (the lower section as far as Liberia)
5. Spain and the upper section of the West African coast
6. Atlantic coasts of Europe and the British Isles

This cavalier approach to west African toponymy strengthens what was said earlier: 'It was clearly not a priority for Benincasa to include the full complement of names - they would have been meaningless to most of his users and only the Portuguese navigated those coasts - and so he continued until he ran out of space, sometimes half obscuring the final name in the margin. In other words, what was shown depended on drafting decisions, not hydrographic ones' (see the Benincasa essay under West African coast (towards the end of the 'Stage 3' section).

I am not aware of any accessible scans of the Budapest atlas, just, it seems, as none are available for the Kiev charts. Until that happens, a full study cannot be made of these two versions. It is unlikely, though, that there will be more than superficial differences between them. Certainly, the title inscription in both 1474 atlases is of the same restrained form, with the earlier long and ornate calligraphic tail - whose faint residual outline can still be made out on the British Library's 1473 atlas - now abandoned.

Where sample images from the Kiev atlas have been made privately available to me I have added the information to the respective Benincasa Tables.

The assiduous Jens Finke has notified me (August 2016) of the following online scans:

• a readable but greyscale image of the Black Sea/Eastern Mediterranean pair from a 1899 publication
• a black and white version of just the Black Sea (click on the image to enlarge to medium resolution)
• a small detail of the Nile delta in a recent article in Ukranian (p.361, comparing the Kiev and Budapest versions)

One trifling feature that I could not find on any of the reproductions of Benincasa's work available to me, and which may therefore be unique, is the treatment on the Kiev atlas's final sheet of the large red inscription over Ireland: IRlanda que Ibernia dicitur. Whereas the main headings for the three component parts of the British islands have, as expected, a large full-stop at either side, in this case dicitur is followed by a colon whose two elements are divided by a prominent horizontal wavy line. The almost identical iterations of his atlases must have involved endless repetition but Benincasa was still capable of introducing small modifications, or perhaps needed to do that for his sanity.

Archives départementales de Vaucluse (Avignon), Port 01, 3E 54, 888bis

An apparently early fragment, without signature or date [which might have been on a missing section], retrieved (damaged) from a notorial register of 1534-5.

The chart covers much of the north-west section of a normal chart, with parts of the North Sea coasts (including England), France and Spain, north-west Africa and parts of the northern Mediterranean and Adriatic. It was discovered in Avignon in 2002 but its importance was first recognised by Paul Fermon in 2015.

The Avignon chart’s dating remains uncertain, despite five years of dedicated research by Jacques Mille. His preferred date of 1310 at the latest (see 2021, p.260) has to be considered alongside the evolving knowledge of the British Isles revealed on the charts of Pietro Vesconte in the 1310s. That the Bristol Channel is indicated on the Avignon chart but not recorded by Vesconte until 1318 is not conclusive evidence of a later date for the Avignon chart since the similar outlines in each case are paired with noticeably different toponymy. In other words, the different sources that seem to have been involved point to parallel development not the clear progress that can be seen on successive Vesconte charts. {This paragraph added 17 March 2022}

For illustrations, a full description of the chart and comparison with others possibly contemporary with it, and with successive summaries of our present understanding of it, see:

Jacques Mille. 'The Avignon Chart (c.1300–c.1310): An Early Attempt to Represent the Northern Regions of Europe on a Nautical Chart,’ Imago Mundi 75, no. 2 (2023).
Jacques Mille. ’Last news about the Avignon Chart’. A YouTube video of a talk to the IVth International Workshop: ‘On the Origin and Evolution of the Nautical Chart’, Lisbon, 25 May 2023 (Session 1, 27-64 mins).
Jacques Mille.La carte d'Avignon: De la Méditerranée à la Baltique 1190-1490 (Espelette: Pyrénées-Atlantiques, 2021).
Jacques Mille. De la Méditerranée à la mer Baltique. 1190-1490. Recherches 2015-2020 sur les cartes marines et les portulans [Privately published, 2021]. [A 361-page, richly illustrated summation of his research into the Avignon chart, with further conclusions about the chart's North Sea coverage.]

Jacques Mille. 'The Avignon chart. Something new in the little family of the first portolan charts' (2018). [72-page, illustrated booklet, examining in detail the recently discovered Avignon chart and proposing a reading of the northern regions (Britain, the North Sea & Baltic) which are drawn very differently than on the other early charts. Self-published at the time of the 2nd Lisbon Workshop (7-8 June 2018), in 60 copies; it can be purchased from the author (jacques.mille2 (at) @wanadoo.fr)].

Jacques Mille. 'The French Mediterranean coasts on portolan charts' (2016). [60- page, illustrated booklet, examining in detail the coastal configuration and toponymy, particularly among the earliest charts, where it is suggested that the Cortona chart might be older than the Carte Pisane. Self-published at the time of the Lisbon Workshop (6-7 June 2016), in 60 copies; it can be purchased from the author (jacques.mille2 (at) wanadoo.fr)].

Jacques Mille & Paul Fermon. 'Une carte portulan récemment découverte. Peut-être une des plus anciennes conservées? La carte d’Avignon. L’histoire d’une découverte', Brussels Map Circle website, '1 May 2017' [but released on the web in September 2017].

Ramon Pujades, in his major new study, refutes the dating to the very early 14th century proposed by Jacques Mille. 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]. 786 pp., profusely illustrated and freely available online HERE, especially on pp. 101 / 123 - 105 / 127.

His conclusion (p.105) is that, "Pending the dating of its parchment that could be determined by a future carbon-14 analysis, everything decisively points towards a cartographic pattern created between 1318 and 1327, that is, in parallel to the cartographic activity carried out by Perrino Vesconte and Giovanni da Carignano". His detailed argument focuses on the treatment of the North Sea coasts and those of eastern England, which he compares with the later Vesconte charts. {This section added 16 September 2024}.

Universiteits Bibliotheek Gent, BHSL.HS.0080/2

A large section of a Catalan chart, c.1440-1460, provisionally attributed to a member of the Soler family.

In February 2021 Joaquim Alves Gaspar discovered a chart fragment in Belgium (measuring 54 x 29 cm). 'Discovered' is perhaps not the right word because this has been in the Ghent University Library since the 19th century, along with another, later fragment. There is no known provenance for those. This welcome addition to the list of surviving charts prior to 1500 is the first for several years.

It is described on the MEDEA-CHART Database, managed by Gaspar at the the Faculty of Sciences, University of Lisbon. Although the chart is worn, water-stained and faded – perhaps pointing to the reason it was discarded and presumably re-purposed – much of it is legible in the enlargeable scan offered on the MEDEA site. That also provides a coverage diagram which shows that what remains comprises the greater part of the original chart, as can be seen from the extent of the compass line network with its centre in Calabria.

As to the chart's dating and attribution, Ramon Pujades has concluded, on the basis of its toponymy and handwriting, that it is clearly a Catalan production (i.e. drawn in Palma de Majorca). It is perhaps datable to the middle years of the 15th century as the product of the Soler family (on which see Pujades's 2009 publication, La carta de Gabriel de Vallseca de 1439 (pp.312-16).

One further observation, albeit one that might seem trivial. There are occasional idiosyncratic features on the portolan charts that can serve as the hidden 'signatures' of their creators. One such was the treatment of two names, Aigues Mortes and Damiata, by separating off their initial letter and placing it on a nearby island. [On this see 'Detached initial letters'.] In the case of Aigues Mortes I had noted only a single instance where, rather than just transferring the 'A', it was the first two letters (ay) that were extracted. And that was found in the chart signed by Rafel Soler (but not dated) preserved at the Humboldt University in Berlin.

Archives départementales du Var (Draguignan).

Two large fragments (3 E 3441 and 3 E 3443) combining to form most of a Catalan chart, recovered from the binding of notorial registers, dating from 1591 to 1597, from the village of Solliès. Emmanuelle Vagnon has suggested that the author might be either Gabriel de Valseca (fl. 1439-49) or Petrus Roselli (fl. 1447-69).

Although considered to form part of a single chart I found it difficult to match up the compass lines. It certainly looks that Roselli was involved, even if only through a copyist of his work. Roselli has been credited, no later than 1456, with doubling the compass lines from 16 to 32, as can be seen here. It also seems that Roselli only introduced the dominant Atlas Mountains and the Alps on his later charts. He regularly included vignettes of both Genoa and Venice but the small surviving portion of the Venice illustration looks quite different to his normal work. A different artist perhaps?

Further detailed analysis, for example handwriting comparison, might clarify if we have been made aware of one or even two further Roselli-style charts. The provisional dating, from the end of the 15th or beginning of the 16th century, might also become clearer.

Literature
Emmanuelle Vagnon, 'Portolan charts in French and German archives. New discoveries'. A YouTube video of a talk to the IVth International Workshop: ‘On the Origin and Evolution of the Nautical Chart’, Lisbon, 25 May 2023 (Session 3: minutes 3-9).

'Var: une carte marine du Moyen Âge reliait un registre de notaire', Revue française de Généalogie, 6 July 2022 (a brief note by Guillaume de Morant).

Illustrations (enlargeable):
3 E 3441
3 E 3443

Stills from the YouTube video above also provide illustrations of the two parts.