Paintings by Jan van Eyck

Exploring Jan van Eyck’s paintings

How many paintings by Jan van Eyck survive? The answer depends on what is being counted. Nine extant pictures bear his signature, according to Lorne Campbell's catalogue. An exhibition at Berlin's Gemäldegalerie describes roughly twenty works as signed or securely attributable to him, with further paintings from the workshop. A large altarpiece is one work made of many panels; a portrait with its original frame is a different sort of object; a copy after a lost composition preserves an image without preserving Jan's paint. The most useful way to explore his paintings is to keep those distinctions visible.^[1][2]

The surviving signed and dated sequence discussed by Campbell runs from 1432 through 1439. Jan was already a working court painter before 1432, but no earlier painting can be placed in a comparable sequence with confidence. Some original frames, where he often wrote his name and date, have disappeared. Pictures known through early descriptions are lost; others survive with uncertain authorship or later intervention. The small body we can inspect today therefore represents a career and a network, not a complete catalogue recovered intact.^[1][2]

Browse Jan van Eyck works

The Ghent Altarpiece

The Arnolfini Portrait

Portrait of a Man (Self Portrait?)

The Rolin Madonna (La Vierge au Chancelier Rolin)

Adoration of the Lamb

Part of the Ghent Altarpiece

The Annunciation (1435)

Annunciation Diptych (Madrid)

One wing of the Madrid diptych

Annunciation Diptych (Madrid)

One wing of the Madrid diptych

The Erythraean Sibyl

Part of the Ghent Altarpiece

The Lucca Madonna

Small Triptych (central panel)

Cardinal Niccolò Albergati (?)

Portrait of Christ

Later copy after a lost composition

God the Father

Part of the Ghent Altarpiece

The Cumaean Sibyl

Part of the Ghent Altarpiece

Portrait of Giovanni Arnolfini

Madonna at the Fountain

Portrait of a Man (Léal Souvenir)

The Madonna in the Church

Portrait of a Man with Carnation

Later copy after Jan van Eyck

St Jerome in his Study

St Francis Receiving the Stigmata (Turin)

St Francis Receiving the Stigmata (Philadelphia)

Portrait of Jan de Leeuw

Portrait of Baudouin de Lannoy

Man in a Blue Chaperon

The Virgin Mary (The Ghent Altarpiece)

Part of the Ghent Altarpiece

The Birth of John the Baptist (Turin–Milan Hours)

Manuscript illumination; attribution debated

The Ince Hall Madonna (The Virgin and Child Reading)

Follower of Jan van Eyck

Donor and St John the Baptist

Part of the Ghent Altarpiece

Virgin and Child with St Michael, St. Catherine and a Donor

The Offerings of Cain and Abel

Part of the Ghent Altarpiece

The Prophet Zacharias and the Angel Gabriel

Part of the Ghent Altarpiece

Isabella Borluut (Ghent Altarpiece)

Part of the Ghent Altarpiece

The Soldiers of Christ

Part of the Ghent Altarpiece

Virgin and Child with Saints and Donor

Jan van Eyck and workshop

Portrait of Margaret van Eyck

Adam (Ghent Altarpiece)

Part of the Ghent Altarpiece

Eve (Ghent Altarpiece)

Part of the Ghent Altarpiece

Madonna and Child with Canon Joris van der Paele

Angel Annunciate (Ghent Altarpiece)

Part of the Ghent Altarpiece

Virgin Annunciate, from the exterior of the right panel of the Ghent Alterpiece

Part of the Ghent Altarpiece

Woman Bathing

Lost painting; surviving witnesses

The Just Judges

Ghent component; replacement for the stolen original

The Three Marys at the Tomb

Attributed to Jan van Eyck

Crucifixion and Last Judgement Diptych

Crucifixion (Diptych Panel)

Last Judgement (Diptych Panel)

Adoration of the Mystic Lamb

Part of the Ghent Altarpiece

Singing Angels

Part of the Ghent Altarpiece

Portrait of Isabella of Portugal

Lost portrait; later visual witness

Music Making Angels

Part of the Ghent Altarpiece

Back Panel Ghent Altarpiece with Interior View

Part of the Ghent Altarpiece

Vera Icon

Later copy after a lost composition

Madonna of Nicolas van Maelbeke

Lost painting; later visual witness

Saint Christopher

Later Eyckian painting

A date written on an original frame and a date proposed by a museum are different sorts of evidence. The former may anchor when an object was completed, provided the frame and inscription are genuinely part of it. The latter often places an undated picture by comparing style, materials and circumstances with better documented works. This is why the 1434 Arnolfini Portrait and 1439 Madonna at the Fountain can appear beside works dated “about” a year or a span of years. A chronological list is useful for orientation, but its apparent precision varies from entry to entry.^[1][2][3][4]

What does remain ranges from the many-panelled Ghent Altarpiece to a devotional image only a few inches high. It includes portraits, pictures of Mary and the Christ Child, narratives of the Annunciation, the Crucifixion and the Last Judgment, and designs that others carried forward. This guide follows the works by their forms and uses. The biography follows the documents of the painter's life; the individual work guides allow a closer view of each object.^[1][2]

How to read dates and attribution labels

Campbell's nine signed pictures make a useful backbone for comparison: Léal Souvenir in 1432, the possible self-portrait in 1433, Arnolfini in 1434, Jan de Leeuw and Van der Paele in 1436, the Dresden triptych and Saint Barbara in 1437, and Margareta and Madonna at the Fountain in 1439. This sequence does not account for every surviving object. It groups works whose identifying inscriptions can be studied, including a panel whose degree of finish is debated. Ghent has its own shared attribution in the frame quatrain.^[1][5]

A catalogue also needs to distinguish a work from a detail and an invention from its execution. The central Lamb scene, its surrounding pilgrims and the figures of the upper register belong to one Ghent commission. The two Metropolitan narratives belong to one linked object history, although each has a separate guide here. By contrast, the two Saint Francis versions are different surviving paintings. Later artists could repeat an Eyckian composition on fresh supports. Those repetitions help reconstruct its reception, but a catalogue of Jan's surviving paint cannot count every one as another autograph work.^[2][6][7]

Authorship labels express different judgments. “Attributed to” records a reasoned proposal; “workshop” allows production within a master's organisation; “after” credits an earlier model while identifying a later maker. These are not interchangeable stages on a simple scale of quality. A posthumous workshop completion can preserve an original commission, while a brilliant later copy belongs to a different historical transaction. The collection examples below show why the qualification should travel with the image when it is reproduced.^[8][9][10]

One altarpiece, many views

The Ghent Altarpiece occupies a special place in any account of Jan's paintings. It was made for the chapel of Joos Vijd and Elisabeth Borluut. An inscription records that Hubert van Eyck began it and Jan completed it for presentation in 1432. The work opens and closes: its exterior includes the donors and the Annunciation, while the interior gathers prophets, saints, angels, Adam and Eve around the central vision of the Lamb. It is a painted structure that changes with handling, not a set of unrelated canvases assembled by a modern catalogue.^[5]

Its scale can mislead a simple count. The Adoration of the Mystic Lamb and the Singing Angels each warrant close attention, but each belongs to the whole. The same is true of Adam, Eve and the donor Isabella Borluut. Their separate guides follow details and questions within a single commission. Jan's relationship to Hubert's earlier work has not been settled panel by panel; conservation research continues to test how the brothers' contributions might be understood.^[5][11]

The altarpiece's present appearance combines original work with later interventions. Its Just Judges original was stolen in 1934 and is represented by a modern replacement. Elsewhere, restoration has removed substantial overpaint. The individual panels have also experienced different physical histories: some wings were sawn through their thickness in Berlin in 1894 to display both painted faces simultaneously. Reading a modern photograph therefore requires attention to the state being shown as well as the figure represented.^[11][12]

Opening and closing also gave the altarpiece a sequence. On the shut wings, the Annunciation occupies a quiet, pale interior above donors and patron saints. Opened, the colour and scale of the heavenly assembly change the encounter. The Lamb panel belongs to that interior revelation, while Adam and Eve stand at the far edges of the upper tier. A photograph of one open panel can give admirable detail yet conceal that movement between the two states. The individual panel pages serve readers who want to inspect a figure closely; the whole-work page restores its place in the structure and in the chapel for which it was made.^[5][12]

The Lamb panel also preserves evidence of successive acts of painting. KIK-IRPA's study identifies a developed composition assigned to Hubert beneath Jan's revisions: the earlier scene already included the altar and Lamb, a landscape and a natural spring. Jan changed the spring to the visible fountain and revised figures and landscape. This is evidence of a painted contribution, not merely a claim that Hubert supplied ideas. It is specific to the investigated layers and cannot divide every panel between the brothers without further study.^[13]

Patrons placed before the sacred

The Madonna with Canon Joris van der Paele shows one patron's connection to a church. The canon kneels before Mary and the Christ Child between Saint George and Saint Donatian. The original frame names him and records a commission in 1434 and completion in 1436. Its inscriptions also refer to chaplaincies founded by the donor. At 124.5 by 160 centimetres without its frame, the panel offered a substantial, lasting presence in the Bruges church of Saint Donatian. Its painted richness served the devotion and memory attached to that place.^[14]

The Virgin of Chancellor Rolin presents a different relationship between powerful patron and holy figures. Nicolas Rolin kneels opposite Mary and the Child in an interior that opens through arches to an expansive garden and city. The Louvre dates the picture broadly to about 1430–35. It considers private prayer and a later memorial function as possibilities, while recognising that the original arrangement is not completely recoverable. The painting's minute city is not an identified portrait of a real place. Jan gave the patron a spacious setting without turning the image into a straightforward record of a room or landscape.^[15]

These two works can be viewed together without assuming that they did the same job. Van der Paele's original frame binds an individual to a named church and a documented foundation. Rolin's setting is more elusive, even though the chancellor himself is known. Both allow a living person to appear near Mary and the Child. The distance between them lies in scale, inscription, likely display and the evidence that has survived. Their own guides examine each composition and its history; together they show how varied a donor image could be.^[14][15]

Rolin's panel has a second painted face. C2RMF's account of its 2022–23 treatment describes imitation jasper on the back, recovered through cleaning and made suitable for display. On the front, darkened varnishes and accumulated overpaint had obscured colour and spatial relationships. Treating the two faces changed more than the brightness of a familiar illustration: it restored the possibility of encountering the panel as an object with a front, back and thickness. The full treatment involved decisions about losses and old repairs, which are documented in the conservation account.^[16]

That evidence is relevant when comparing Rolin with Van der Paele. The former measures about 66 by 62 centimetres; the latter is 124.5 by 160 centimetres without its frame. A screen can make them look similarly sized. Their physical scale, documented inscriptions and possible handling imply different encounters. Neither a donor's presence nor a shared subject is enough to reconstruct the first display. The Louvre considers a mobile prayer image followed by a memorial role possible for Rolin; Van der Paele's frame explicitly names the church foundation.^[14][15]

Small Madonnas and private attention

Jan could make the Virgin and Child feel close without relying on a donor's likeness. The Lucca Madonna places them in a narrow domestic interior. The Städel Museum describes a floor, side walls and vaulted ceiling that seem to extend toward the viewer; a window reflected in a glass vessel makes the space tangible. It dates the work to about 1437 and catalogues its medium as mixed technique on oak. This is a reminder that the phrase “oil painting” is sometimes too broad to stand in for the material description of every object.^[17]

The Madonna at the Fountain is smaller still, 19 by 12 centimetres. Mary holds the Child in front of a cloth supported by angels, with a metal fountain in the foreground. Its original marbled frame carries Jan's signature, motto and the date 1439. The museum regards its scale as well suited to private devotion but does not know the first owner. A small panel can retain exact physical evidence about its maker while leaving its patron and first room unidentified.^[3]

KMSKA notes another question about the Fountain panel: its uneven thickness might mean that it was later sawn through, perhaps removing a painted reverse. No documentary record confirms that proposed reverse. The museum also records early versions of the composition and a possible connection with an inventory of Margaret of Austria. These are separate strands of evidence. A repeated design can explain later recognition without identifying the first owner, and a suggestion about the back must remain distinct from the visible, signed front.^[3]

Other Marian works take the intimacy of a small format in different directions. In the Madonna in the Church, an impossibly large Mary stands within Gothic architecture. In the Dresden triptych of 1437, Virgin and Child with Saints Michael and Catherine, the donor appears on the left wing. Dresden describes the small format as intended for private devotion. Painted wings and an exterior Annunciation are parts of one hinged object.^[18] These pictures should not be flattened into a single “Madonna type.” Scale, framing and the arrangement of sacred space change how each would have been encountered.^[1][2]

The Berlin museum describes the Madonna in the Church as a tiny panel, 31.1 by 13.9 centimetres, in which Mary surpasses the size of the building that encloses her. Light enters through windows and a side portal; singing angels appear in the distant choir. The museum interprets this as a sacred vision rather than a portrait of a particular church. The scale reverses an ordinary expectation: a physically small work makes Mary overwhelming within its pictured space. The Lucca Madonna draws a viewer toward an intimate room, while the Berlin panel turns a church interior into a vision. Both ask the viewer to look closely, but their spatial effects are markedly different.^[17][19]

The Dresden triptych adds the practical possibility of a changed view. Open wings bring its saints and donor into relation with the central Virgin and Child; the closed exterior presents the Annunciation. That arrangement gives two subjects to one portable structure. It is consequently misleading to count its central panel, wings and outer faces as separate surviving paintings simply because they can be illustrated separately. The question “how many works?” depends partly on whether the unit is a painted surface, a panel, a hinged object or a commission. The same counting problem is grander at Ghent, but it is visible in this much smaller devotional work too.^[1][2]

Watch: Jan van Eyck in context

Choose a Smarthistory close reading of the Arnolfini Portrait or the closed and opened Ghent Altarpiece.

Jan van Eyck, The Arnolfini Portrait

Smarthistory

Look closely at the room, mirror and gestures in the London double portrait.

Watch on YouTube

▶01 · PortraitThe Arnolfini PortraitSmarthistory▶02 · ExteriorThe Ghent Altarpiece, closedSmarthistory▶03 · InteriorThe Ghent Altarpiece, openSmarthistory

Three different Annunciations

The Washington Annunciation takes place within a church. Gabriel and Mary speak in gold lettering; Mary's answer is turned as if toward heaven. Architecture and painted words participate in the event. It is a separate panel in the National Gallery of Art, not a detail cut from the Ghent Altarpiece.^[20]

The Madrid Annunciation Diptych is a different, paired object. Gabriel and Mary occupy separate wings and appear almost like stone sculptures in niches. The Thyssen-Bornemisza Museum dates the diptych to about 1433–35, despite the later date embedded in the inherited web address. Painted frames carry words from their exchange. The black, white and grey treatment makes the act of imitating sculpture part of the painter's design. Its two wings belong together; neither is a second independent Annunciation by Jan.^[21]

The third setting is the Annunciation across Ghent's closed exterior. Its angel and Virgin form a dialogue within the polyptych's larger sequence. Grouping these three works under one subject helps comparison, but it must not merge their identities. Washington is a painted church scene; Madrid is a compact grisaille diptych; Ghent's scene appears when the great wings are shut. The difference concerns each object's structure and first viewing, not merely its current location.^[5][20][21]

Faces, names and uncertain identities

The Arnolfini Portrait is the best known of Jan's portraits, partly because it makes a room and its objects so persuasive. It is signed and dated 1434. The man is probably Giovanni di Nicolao di Arnolfini, an Italian merchant in Bruges, and the woman is probably his wife; her name is unknown. The National Gallery no longer presents the scene as a proven wedding. Its old shutters are lost, so the present painting is also an altered version of the object that earlier owners handled. Those qualifications leave a richly observed portrait rather than a solved anecdote.^[4]

Other sitters invite different kinds of certainty. The National Gallery's 1433 Portrait of a Man may show Jan himself, but the painter's identity is an interpretation, not a fact supplied by the original frame. In the 1432 Léal Souvenir, a dated inscription and a carefully painted stone parapet survive while the sitter's name does not. The person appears vividly present yet remains historically unidentified. Both portraits ask us to separate skill in likeness from the later recovery of a biography.^[22][23]

The portrait of Margareta van Eyck makes the opposite case. Its original marbled frame identifies her as Jan's wife, gives her age as thirty-three and dates the picture to 1439. The reverse is painted like marble as well. Name, date and physical construction work together here; a reader can make a firmer claim about her than about the Léal Souvenir sitter. The portrait of Jan de Leeuw likewise retains identifying words for a Bruges goldsmith, with a little lion standing for his surname.^[24] An original frame can be a source of evidence as important as the face it encloses.^[1][25]

Not every traditional name remains equally secure. The portrait customarily called Cardinal Albergati has been identified and dated in different ways; the related portrait drawing raises its own questions. The Man in a Blue Cap has a proposed sitter rather than a universally accepted one. A guide can preserve familiar titles so works can be found, while the individual records test what those titles actually prove. The same care applies to Baudouin de Lannoy and the separate Berlin portrait of Giovanni Arnolfini, which must not be collapsed into the London double portrait.^[1][2]

The Vienna museum itself prints a question mark beside the name of Cardinal Niccolò Albergati. It connects the portrait with a silverpoint drawing in Dresden, where notes record the colours of the sitter's face and clothing. The drawing is unusually useful evidence of looking and preparation, yet it does not turn every aspect of the later painted identity into certainty. It is also a separate surviving object. Comparing the two can reveal how an artist moved from a small observed study toward a painted likeness; adding both to a count of “paintings” would obscure that distinction.^[26]

Narratives in small panels

The Metropolitan Museum's Crucifixion and Last Judgment sets two dense scenes side by side. The Met currently catalogues the paintings to Jan van Eyck and dates them about 1436–38. Older publications and the museum's 2016 exhibition used about 1440–41; the difference matters when placing them in his career. Technical research indicates that the panels originally served as wings of a triptych or doors of a tabernacle or reliquary, rather than as the independent diptych familiar today. Their precise first setting remains unresolved. The separate Crucifixion and Last Judgment guides are views into the pair, not additional independent paintings.^[6]

The Metropolitan pair's frames are original, although the pictures themselves were transferred from wood to canvas. Their raised Latin inscriptions are accompanied by fragmentary Middle Dutch text found beneath later gold paint. The 2016 investigation used X-radiography to reveal that hidden wording and infrared reflectography to compare the painted underdrawings with the Rotterdam sheet. Frame, support and preparatory image each supply a different part of the evidence for the lost arrangement. “Diptych” is a convenient way of finding the present pair, but not a complete account of its original construction.^[6][27]

A Crucifixion drawing in Rotterdam is related to that discussion but remains a work on paper. Its relation to the Metropolitan panels is one reason scholars compare underdrawing, painted revisions and frame evidence. A matching subject does not turn a sheet into a painted panel or establish a simple sequence from one to the other. The object type matters whenever the same image moves across media.^[6]

Two small paintings of Saint Francis receiving the stigmata in Turin and Philadelphia present a similar subject in separate surviving objects. The Philadelphia museum dates its panel about 1430–32 and stresses its portable scale. It shows the saint's visionary encounter amid rock and distant city. The pair invites questions about versions and the circulation of a design, while each panel requires its own material and provenance record. Neither should disappear under one generic “Saint Francis” entry.^[28]

The two Saint Francis images were brought together for Recognizing Van Eyck, organised by Philadelphia and the Galleria Sabauda in collaboration with the National Gallery, London, in 1997–98. The exhibition treated their close resemblance as a reason for joint examination. A viewer can compare the sleeping companion, the saint's position and the rocky setting, then return to each object's materials and history. Repetition of the design does not establish identical circumstances of production, nor does it make one physical painting disappear from the surviving corpus.^[28][7]

The Saint Jerome in His Study in Detroit places a scholar-saint among books, an astrolabe and an hourglass. The museum currently catalogues it to Jan, about 1435, on a complex support described as oil on linen paper on oak. A possible date inscribed in the picture and the history of past attributions show why even a current museum label should be read alongside technical and documentary evidence. The panel adds another kind of intellectual interior to the surviving range without making its attribution history disappear.^[29]

What the material can tell us

Jan's reputation for astonishing surfaces is well deserved, but “realism” is too small an account of how the pictures were made. Oil-based paint allowed translucent layers, gradual transitions and precise highlights; it had been used before Jan, so he did not invent the medium. Different supports and techniques appear among the works in this guide: the Städel calls the Lucca Madonna mixed technique on oak, and Detroit describes its Saint Jerome as oil on linen paper on oak. Each museum's object record is a better guide to a particular painting than a single statement about all of them.^[17][29][30]

The Saint Barbara makes the question of finish visible. Dated 1437 on its original frame, it gives a tower under construction extraordinary definition while using little colour. The KMSKA reports continuing disagreement: was it abandoned before completion, or intended as a restrained grisaille? Even its frame bears signs of incomplete marbling. Calling it simply an “unfinished painting” turns a live interpretation into a settled object fact. The panel asks viewers to attend to drawing, selective colour and the material state of a work that sits at the edge of neat categories.^[31]

Frames can change what counts as evidence. The Van der Paele and Margareta inscriptions preserve patrons, dates and relationships. The Madrid diptych makes words and simulated stone part of the scene. The Met pair's buried frame text shows that later repainting can conceal original information. At Ghent, conservation has also exposed passages long hidden under later additions. A catalogue entry records the object as it is understood now; conservation may refine its attribution, date or original appearance. The physical painting therefore remains a source to be questioned, rather than a static illustration of a title.^{[11][14][21][25][6]}

Condition also changes the evidence available for reconstruction. In the Rotterdam Three Marys, the museum's Dutch-language curatorial response explains that too little survives to establish whether the panel was once larger on both sides or only one. It therefore declines to reconstruct the missing whole. The same source identifies the Hebrew-like decoration on the women's robes as pseudo-writing. An apparent inscription may be ornament rather than a legible historical message, just as an apparent complete view may belong to a cropped object.^[8]

Attribution, copies and lost originals

The painter's name travelled through his workshop and through later admiration. The Rotterdam Three Marys at the Tomb is attributed by Museum Boijmans Van Beuningen to Jan, while the museum explicitly leaves open whether other hands contributed. The Frick's Virgin and Child with Saints and Donor, commissioned by the Carthusian Jan Vos near the end of Jan's life, is described as Jan and workshop; an assistant apparently completed it after his death. These are neither simple autograph panels nor mere later copies. Their labels preserve what can be said about design and execution.^[8][9]

Jan Vos's commission offers a particularly clear reason to keep these labels precise. The Berlin museum's account says Jan may have begun the larger panel before his death in 1441, but workshop assistants executed most of it. The image was in place for worship in Bruges by 1443 and later moved with Vos to Utrecht. A smaller painting by Petrus Christus in Berlin freely reworks part of the same composition for a different setting. The chain is therefore more specific than a vague claim of “influence”: a patron commissioned a workshop picture, carried it to another house, and obtained a related image from another painter. The Frick panel, its workshop completion and the Berlin derivative each have distinct authorship, use and material history.^[9][32]

Other images show what a copy can and cannot offer. Berlin catalogues its Man with the Pinks as the work of an unknown painter after Jan, made around 1520. The Vera Icon in Bruges is an anonymous early seventeenth-century copy after a lost Eyckian prototype; its inscription credits Jan as inventor and gives 30 January 1440 for the model. The extant panel is much later than that written date. The Ince Hall Madonna is treated in the National Gallery of Victoria's reassessment as a follower's later imitation. Its examination history includes arguments for returning it to Jan or his studio, but the museum explains why it retains the follower attribution. These works have value as witnesses to transmission. Their makers and surfaces are their own.^[10][33][34]

Some works have no surviving original panel to inspect. Jan's portrait of Isabella of Portugal is documented through his 1428–29 mission, but the painting sent to Philip the Good has disappeared. The Madonna of Nicolas van Maelbeke is known through later witnesses to a lost memorial composition. KIK-IRPA identifies a lost Woman Bathing by Jan among the pictures represented on the wall in Willem van Haecht’s Gallery of Cornelis van der Geest of 1628.^[35] The reader can learn about patronage, subjects and reception from such witnesses without pretending to look at a panel that no longer exists.^[1][9]

Maelbeke's lost composition illustrates how those witnesses differ. A silverpoint drawing attributed to Petrus Christus, shown by the Frick, records the Virgin and architectural setting while leaving out much of the landscape and donor group. It is a selective response rather than a complete facsimile. The Met also holds a 1575–85 engraving from the circle of the Wierix brothers, catalogued “after Jan van Eyck.” Comparing these sources can trace transmission across generations and media, while neither permits inspection of the lost original's surface.^[36][37]

The lost Isabella portrait has a different evidential basis. Bart Fransen's study follows the embassy account that states Jan portrayed Philip the Good's prospective bride in Portugal in 1429. The act of making a likeness is documented even though the painted likeness has vanished. It consequently belongs in a record of the painter's output, but not among the objects a visitor can see as surviving Jan panels. The Isabella page distinguishes the documentary mission from later images associated with her name.^[38]

One route in the wider group concerns a different medium altogether: The Birth of John the Baptist belongs to the Turin–Milan Hours manuscript. Its “Hand G” attribution has been debated, and a scholarly reassessment rejects Jan's authorship. It is not a panel of the Ghent Altarpiece. The drawings and prints guides likewise distinguish original graphic work from later reproductions after paintings. A broad browse list is most useful when its headings state these boundaries clearly.^[1][39]

The Saint Christopher route also belongs to this wider history. Philadelphia dates its small painting about 1460–70, after Jan's death, and relates its detailed execution to his school. A separate Louvre drawing, dated about 1480 and catalogued as a copy after Jan, supports discussion of a possible lost Eyckian invention. The Louvre records disagreement over whether the whole design or only the saint derives from that model. The later panel, the drawing and the proposed lost painting must therefore retain separate identities.^[40][41]

Choosing where to look next

For the widest change of scale, compare the Ghent whole with the tiny Antwerp Fountain and Philadelphia Saint Francis. For two ways of placing a donor before Mary, compare Van der Paele and Rolin. For painted sculpture, turn from the colourful Washington church to the pale Madrid figures. For the evidential value of a frame, put Margareta beside the unidentified Léal Souvenir sitter. These comparisons begin with identifiable properties of the objects rather than a ranking of famous names.

The artwork grid provides routes into the individual guides. A panel title may identify one part of a larger work; a familiar traditional name may remain uncertain; a later copy may preserve a composition that is lost. The linked articles retain those distinctions while examining each object's making, use, condition and ownership. The drawings and prints guides follow the related works in other media, and the biography places the documented commissions within Jan's working life.

Related reading

Ghent Altarpiece

Adoration of the Mystic Lamb — Jan and Hubert van Eyck

The Singing Angels of the Ghent Altarpiece

References

1. Lorne Campbell, “Jan van Eyck,” in The Fifteenth Century Netherlandish Schools (London: National Gallery Publications and Yale University Press, 1998; online version 2025). View Source.
2. Staatliche Museen zu Berlin, Jan van Eyck, bilingual exhibition text. View Source.
3. KMSKA, “Madonna at the Fountain,” collection record 411. View Source.
4. National Gallery, London, “The Arnolfini Portrait,” collection record NG186. View Source.
5. Saint Bavo Cathedral, “The Ghent Altarpiece,” history and commission record. View Source.
6. Metropolitan Museum of Art, “The Crucifixion; The Last Judgment,” 33.92ab, current object record: Jan van Eyck, ca. 1436–38, transferred support and original frames. View Source.
7. Philadelphia Museum of Art, “Recognizing Van Eyck,” exhibition record (1998). View Source.
8. Museum Boijmans Van Beuningen, “The Three Marys at the Tomb,” collection record. View Source.
9. Frick Collection, “The Charterhouse of Bruges: Jan van Eyck, Petrus Christus, and Jan Vos,” research article. View Source.
10. Staatliche Museen zu Berlin, “Der Mann mit den Nelken,” collection record 525A. View Source.
11. KIK/IRPA, “Restoration and research of the upper register of the interior of the Ghent Altarpiece.” View Source.
12. Saint Bavo Cathedral, panel guide: closed/open programme and the stolen Just Judges. View Source.
13. KIK-IRPA, “The mystery of Hubert van Eyck and the Ghent Altarpiece solved”: lower-register painted stages. View Source.
14. Musea Brugge, “Madonna with Canon Joris van der Paele,” collection record 0000.GRO0161.I. View Source.
15. Musée du Louvre, “La Vierge du chancelier Rolin,” collection record INV 1271. View Source.
16. C2RMF, “La Vierge du chancelier Rolin — Restauration,” 20 March 2024: treatment, painted reverse and display. View Source.
17. Städel Museum, “Lucca Madonna,” collection record 944. View Source.
18. Staatliche Kunstsammlungen Dresden, “Triptych of Mary and Child, St. Michael, and the Catherine,” institutional collection text on Google Arts & Culture: 1437, private devotion and donor. View Source.
19. Staatliche Museen zu Berlin, “Die Madonna in der Kirche,” collection record 525C. View Source.
20. National Gallery of Art, Washington, “The Annunciation,” collection record. View Source.
21. Museo Nacional Thyssen-Bornemisza, “Díptico de la Anunciación,” collection record 137.b. View Source.
22. National Gallery, London, “Portrait of a Man (Self Portrait?),” collection record NG222. View Source.
23. National Gallery, London, “Portrait of a Man (‘Léal Souvenir’),” collection record NG290. View Source.
24. Kunsthistorisches Museum, “Schaulust und Sinnestäuschung”: Jan de Leeuw portrait, 1436, and the painted frame. View Source.
25. Musea Brugge, “Portrait of Margareta van Eyck,” collection record 0000.GRO0162.I. View Source.
26. Kunsthistorisches Museum, Vienna, “Cardinal Niccolò Albergati (?),” collection record 975. View Source.
27. Metropolitan Museum of Art, “A New Look at a Van Eyck Masterpiece,” 2016 exhibition research: original configuration, frame texts and infrared comparison. View Source.
28. Philadelphia Museum of Art, “Saint Francis of Assisi Receiving the Stigmata,” collection record 102076. View Source.
29. Detroit Institute of Arts, “Saint Jerome in His Study,” collection record 25.4. View Source.
30. Susan Jones, “Painting in Oil in the Low Countries and Its Spread to Southern Europe,” Metropolitan Museum of Art, Heilbrunn Timeline of Art History (2002). View Source.
31. KMSKA, “Saint Barbara,” collection record 410. View Source.
32. Staatliche Museen zu Berlin, “Die Madonna mit der heiligen Barbara und dem Kartäuser Jan Vos / Exeter-Madonna,” collection record. View Source.
33. Musea Brugge, “Vera Icon,” collection record 0000.GRO0206.I. View Source.
34. National Gallery of Victoria, “The National Gallery of Victoria's Virgin and Child, by a follower of Jan van Eyck: a continuing reassessment.” View Source.
35. KIK-IRPA, “The Gallery of Cornelis van der Geest gets a new lease of life”: the 1628 gallery picture as a witness to the lost Woman Bathing. View Source.
36. Frick Collection, “The Virgin and Child with a Donor,” Charterhouse exhibition: silverpoint witness attributed to Petrus Christus after the lost Maelbeke composition. View Source.
37. Metropolitan Museum of Art, “Virgin and Child in a Church, also known as the Van Maelbeke Virgin,” 53.601.18(192): circle of Wierix, 1575–85, engraving after Jan. View Source.
38. Bart Fransen, “Jan van Eyck y España. Un viaje y una obra,” Anales de Historia del Arte 22 (2012), pp. 39–58, especially 40–42: documented lost Isabella portrait. View Source.
39. Carol Herselle Krinsky, “The Turin-Milan Hours: Revised Dating and Attribution,” Journal of Historians of Netherlandish Art 6, no. 2 (2014). View Source.
40. Philadelphia Museum of Art, “Saint Christopher,” collection record 102108, ca. 1460–70: later painting associated with Van Eyck’s school. View Source.
41. Musée du Louvre, “Saint Christophe passant l’Enfant Jésus,” INV 20664, ca. 1480: drawing after a proposed lost Eyckian composition. View Source.