Virgin and Child in a Niche

The Artist: For a biography of Jan van Eyck, see the Target=”_blank”>Heilbrunn Timeline of Art History.

The Painting: Lovingly embracing her child, the Virgin stands before a resplendent brocade cloth of honor and throne within a Gothic niche. The architectural setting serves as a shrine or an aedicule, symbolic of the institutional Church, presenting the Virgin not only as the Mother of God but also as the representative of Christians who comprise the Church.[1] Suspended from the buttresses above the Virgin, and extending down beneath her feet, is a cloth of honor in blue silk brocaded with gold. On the upper border of the green velvet canopy, embroidered with pearls, is a Latin text from Genesis 28:17: DOMVS.dEI.EST.ET.PORTA.C[O]ELI, ([This is none other than] the house of God, and [this is] the gate of heaven). It refers to the Virgin as the Porta Coeli (the gate or door of heaven), signaling her intercessory role for humankind through Christ, her son, at the Last Judgment.[2] Fictively carved into the dais below is another Latin phrase from Genesis 24:44: IPSA EST [MVLIER] QVAM PR[A]EPARAVIT DOM[INV]S FILIO D[OMI]NI MEI (Let the same be [the woman] whom the Lord hath appointed out for my master’s son). Both Genesis texts are selections from the story of Rebekah and her son, Jacob, indicating their individual roles as chosen by God. Genesis 24:44 refers to God’s choice of Rebekah as the wife of Isaac, who was the son of Abraham, a leader of the Israelites, and in Christian belief, a precursor of Christ. Genesis 28:17 denotes the divine selection of Jacob as the head of the tribes of Israel.[3] These Old Testament figures are a prefiguration of God’s choice of Mary to be the Mother and mystical Bride of Christ, the latter role of which is signaled by the ring she wears on her left hand. Moreover, exegetical literature on the Song of Solomon (Canticle of Canticles), early on from Origen and Augustine until the twelfth century, equated the bride therein with the Church and the bridegroom to Christ.[4] Rupert of Deutz (ca. 1075/80–ca. 1129) published seven books of commentary on the Canticle where he associated the bridegroom with Christ and the spouse with the Virgin Mary.[5]

The niches to either side of the Virgin and Christ contain sculptures of the Old Testament prophets Moses and Isaiah (?), at the top left and right respectively. At the lower left is an allegorical figure of Ecclesia (the Church), holding a cross and a chalice as her attributes, while at the lower right another figure represents Synagogue, shown blindfolded and carrying the tablets of the Law. The representation of these two figures signifies the triumph of the Church over the Synagogue and over the Jews who do not recognize Christ.

This is also the theme of the far larger and more detailed Fountain of Grace, produced by the workshop of Jan van Eyck about 1440–50 (Museo Nacional del Prado, Madrid; see fig. 1 above), and a later sixteenth-century copy, attributed to Cristóbal de Velasco (Allen Memorial Art Museum, Oberlin, Ohio).[6] Although the upper part of the Gothic tower over the Virgin in The Met’s painting was cut off at an unknown date, what remains of the architecture is reminiscent of the central Gothic tower over Christ in the Prado painting with lateral niches filled with sculptures. Christ, like the Virgin, inhabits a niche, enthroned before a cloth of honor. Looking more closely, the Virgin’s cloth of honor in blue silk brocaded with gold is identical to the pattern of the red velvet brocaded with gold behind the figure of John the Evangelist in the Fountain of Grace (Bruyn 1957, figs. 2–3). In a certain sense, The Met’s painting reduces the elaborate presentation of the Prado painting to the essential message of the Church triumphant over the Synagogue, but here with the Virgin and Child as its theological focus.[7]

There was an ardent devotion to the Virgin in Flanders during the late fifteenth century, due in part to the region’s many Marian relics that became an impetus for a wide variety of depictions of her in all media. The affectionate embrace of the mother and child in the present example is ultimately derived from Byzantine icons that circulated in the Low Countries, imported during the fifteenth century when they were installed in churches and collected by the dukes of Burgundy and the nobility.[8] The pose of the Virgin and Child is known as the highly popular Eleousa type, or Virgin of Tenderness, which was produced in Byzantine mosaics and panel paintings (see The Met, 2008.352).[9] Early Netherlandish painters did not copy these examples exactly. Instead, they adapted the static, iconic presentation of Byzantine icons to more naturalistic flesh and blood representations of human beings. Jan van Eyck led the way with his particular reformulation of the Eleousa Virgin in the Virgin and Child at the Fountain of 1439 (Koninklijk Museum voor Schone Kunsten, Antwerp, fig. 4), the figures of which are closely copied in the present painting -- that is, except for the left arm of the Child that embraces the Virgin rather than hanging down to the side, holding a rosary, and the red rather than blue robe of the Virgin. Perhaps inspired by a particular Byzantine icon, Van Eyck’s rendition became popular in its own time and beyond. The entire painting as well as the motif of the Virgin and Child were produced in fifteenth-century copies by followers of Van Eyck locally and abroad in France and even Southern Germany (figs. 5-7).[10] Still popular in Van Eyck’s Bruges in the early sixteenth century, the motif of the Virgin and Child was incorporated in paintings by artists such as Gerard David (see The Met, 1977.1.1 and fig. 8) and Adriaen Isenbrant (Private collection, fig. 9). Patrons may have specifically requested the copies, hoping for the trickle-down effect of the perceived benefits—whether curative for the spirit or for particular afflictions of the body—that devotion to these images could provide. Perhaps indulgences were attached to some of these modern-day icons, thus providing the promise of an individual’s delivery to Paradise from a lengthy stay in purgatory.

The Attribution and Date: The Met’s painting was first attributed to Jan van Eyck because of the near-identical figures found in his Virgin and Child at the Fountain of 1439 (Nieuwenhuys 1843; fig. 4). By the end of the nineteenth century, it was thought to be by Van Eyck’s closest follower in Bruges, Petrus Christus (Kaemmerer 1898), although Friedländer (1924) regarded it as falling stylistically between the two artists. The association with the works of Van Eyck is understandable not only for the close copy of the motif of the Virgin and Child from the Antwerp painting, but also for the relationship with the ecclesiastical architecture and pattern of John’s cloth of honor, as well as the iconography of the Prado Fountain of Grace.

Contemporary variants on the popular theme of the Virgin and Child in a Niche are found in Rogier van der Weyden’s oeuvre, namely the Standing Madonna of about 1430–32 (Kunsthistorisches Museum, Vienna; fig. 10), the Virgin and Child Enthroned of about 1433 (Museo Nacional Thyssen-Bornemisza, Madrid; fig. 11), and the Durán Madonna of about 1435–38 (Museo Nacional del Prado, Madrid; fig. 12). The Met’s painting shares with these examples not only the general idea of the Virgin and Child in a niche but also the architectural motifs appearing in Rogier’s paintings, although not in the specific arrangement of forms exhibited therein. Perhaps also closely connected to the design of The Met’s painting is an Eyckian drawing of an Annunciation (Herzog August Bibliothek, Wolfenbüttel, Cod. Guelf. Extravaganza 210, f. 472; fig. 13), at the left depicting the pillars of the church’s exterior that are similar to the details of The Met’s painting (Bruyn 1957, pp. 124–26, Borchert 2016, p. 84; 2018, p. 111). Such architectural models were available to painters through prints, such as one by the Master WA, or through other paintings featuring such details (see Anonymous, Architectural Fragment, about 1460–80, Musées des Arts Décoratifs, Paris; Borchert 2018, pp. 110–11). The top portion of The Met’s painting has been cut and can be determined to be missing about 8.5 cm (see Technical Notes). Thus, the side spires likely would have appeared complete, but the central gable would still have been truncated. Perhaps this portion was augmented by framing tracery elements.

Although The Met's painting was previously thought to date around 1500 (see Technical Notes), the composition, architectural motifs, and specific details of the brocade pattern suggest a date earlier in the fifteenth century. Moreover, there are notable stylistic connections when The Met’s painting is compared to the Van Eyck workshop copy of the Virgin and Child at the Fountain of about 1440 (Private collection, New York; fig. 5).[11] Although not painted by the same artist, the two works share the same Eyckian motif of the Virgin and Child (except for the proper left arm of the child) and are stylistically so similar that they may have originated from the same workshop around the same time (compare figs. 5, 14). Both The Met and the private collection paintings were made with semi-integral frames, the lateral members integral with the painted panel and the top and bottom members engaged (see Technical Notes). Both show remnants of greenish paint associated with the frames, The Met’s painting in particular revealing two layers of greenish paint, a lighter green followed by a more translucent green, suggesting perhaps an imitation of a green stone, a finish similar to those we find in Van Eyck and Petrus Christus original painted frames. The flesh areas are thinly applied with warm glazes and selective use of lead white, and in both paintings the heads and hands were emphasized with contours painted with a darker brown paint. Similarly, highlights were worked in on the strands of the Virgin’s hair and in the golden brocade cloths of honor in a manner that imitates Van Eyck’s studied rendering of how light models form and is reflected by different materials. Given all of these connections with Eyckian models, style, and technique, the Virgin and Child in a Niche most likely was produced in close association with Van Eyck’s workshop in Bruges about 1440–50, at the end of Van Eyck’s life or soon thereafter.

Maryan W. Ainsworth 2023

[1] Carol Purtle, The Marian Paintings of Jan van Eyck, Princeton, 1982, p. 141.
[2] Purtle 1982, p. 7. See also, Carla Gottlieb, “’En ipse stat post parietm nostrum’: The Symbolism of the Ghent Annunciation,” Bulletin des Musées Royaux des Beaux-Arts de Belgique 19 (1970), pp. 79ff.
[3] Christian Heck (1996) noted that the text on the canopy (Genesis 28:17) also relates to Jacob’s dream and of the ladder that rises from earth to heaven, supporting the angels of God who descend and ascend on it. In his sermon De beata Maria Virgine, Saint Bernard writes: “She is the ladder…the door…She is Jacob’s ladder…By these rungs the angels go up, and men ascend” (Patrologia Latina, vol. 184, col. 1016).
[4] Purtle 1982, p. 105 and passim.
[5] Rupert of Deutz, Commentaria in Cantica, in Patrologiae cursus completes, series Latina, ed., J. P. Migne, Paris, 1844–55, vol. CLXVIII, pp. 839–963.
[6] See José Juan Preciado, La Fuente de la Gracia, Una Tabla del Entorno de Jan van Eyck, Madrid, Museo Nacional del Prado, 2018.
[7] Pierre Quarré proposed that an intriguing architectural fragment (Paris, Musée des Arts Décoratifs) in the richly ornamented Gothic style was originally part of a painting depicting the Virgin and Child, which he alleged was installed in one of the ducal chapels of the Chartreuse of Champmol in Dijon (Quarré 1957, p. 62). Although the architectural fragment and the setting of The Met painting are not closely related, he suggested that the fragment could have come from a painting that looked something like The Met’s Virgin and Child in a Niche. See also, Stephan Kemperdick in “Jan van Eyck’s ‘Madonna in a Church’ and its Artistic Legacy,” in Maximiliaan Martens et al., Van Eyck, An Optical Revolution, Ghent, 2020, pp. 260–83, esp. pp. 271–73.[8] Maryan W. Ainsworth, “À la façon grèce: The Encounter of Northern Renaissance Artists with Byzantine Icons,” in Helen Evans, ed., Byzantium: Faith and Power (1261–1557), New York, The Metropolitan Museum of Art, 2004, pp. 545–55.
[9] Ainsworth 2004, pp. 578–93; Katrien Lichtert, “The Mother God of Tenderness: The Introduction and Dissemination of an Iconic Image in Early Netherlandish Art,” in Bernard Coulie, ed., Paths to Europe: From Byzantium to the Low Countries, Milan 2017, pp. 116–29; see also Larry Silver, “Eyckian Icons and Copies,” in Magdalena Bellavitis, ed., Making Copies in Europen Art 1400-1600: Shifting Tastes, Modes of Transmission, and Changing Contexts, Leiden and Boston, 2018, pp. 130–64.
[10] For the South German painting, see entry by Stephan Kemperdick in Katrin Dyballa and Stephan Kemperdick, Netherlandish and French Paintings 1400–1480 in the Gemäldegalerie Berlin, forthcoming.
[11] The technical examination of the private collection painting is discussed in Livia Depuydt-Elbaum and Paul Vandenbroeck in “Jan van Eyck Madonna at the Fountain,” Restoration, Koninklijk Museum voor Schone Kunsten, Antwerp, vol. 2, no. 1, 2002, pp. 18–24.

Support: The panel was constructed from two boards of vertically grained oak, with the join located at the center of the panel. The boards were originally held in place with three dowels, as can be seen in the x-radiograph (see fig. 15 above). The oak originated in the Baltic/Polish region.[1] Measurements of the thickness of the panel, which include a thick waxy layer on the back, range from 1 cm at the edges to 1.5 cm at the center. The panel appears to retain its original reverse, although the presence of the wax complicates observations. At some point before entering the museum’s collection, the original panel was trimmed by an unknown amount; estimating the extent is aided by observations about the panel construction and the painted composition.

The painting retains its original edges at the bottom, left, and right. There is a barbe along the bottom edge as well as a margin of wood, originally unpainted but now painted grey, evidence of the removal of an engaged frame member. At the left and right edges there are also wood margins, originally unpainted but now grey, that are slightly proud of the painted surface and flat. There are no barbes, instead the painted surface curves up to the unpainted wood margins. It appears that the frame was integral on the left and right sides, that is, carved from the same piece of wood as the painted panel. There is no barbe or unpainted margin remaining at the top edge of the painting; the panel was cut, cropping the painted composition as well.

The panel seemingly had a semi-integral frame originally, with the lateral members integral to the painted panel and the top and bottom members engaged (fig. 16). To remove an integral frame, the members would have been cut or planed off, slicing through the lip of paint and ground present at the edge of the painting but not leaving a barbe, as appears to have occurred on the left and right sides of this painting. Diagonal cuts at the lower corners on the front and back indicate the location of the miters, and the slight recess on the reverse shows exactly where the engaged member was slotted on to the panel (fig. 17) The engaged members would likely have been held in place with wooden plugs, of which there is no trace now. However, it seems that the unpainted margins were all trimmed somewhat after the original frame was taken off, removing the plug holes on the left and right, as much as 3–4 cm on each side.

The locations of the dowels, as can be seen in the x-radiograph, offer evidence of the extent that the panel was trimmed at the top (fig. 15). Typically, dowels were evenly spaced along the span of a join, but the topmost dowel is located very close to the top of the panel. The topmost dowel is only 5.5 cm from the top of the panel, while there is a spacing of 17.2 and 20.4 cm between the next two, and another 16 cm to the bottom of the panel (keeping in mind that the bottom unpainted margin has been trimmed somewhat).

Using the distance between the lowermost dowel and the bottom edge of the painted composition, 14 cm, we can extrapolate that the painting may have been trimmed by about 8.5 cm on the top. Considering the painted composition, this would not provide sufficient space to complete the central tower in the background of the painting but would place the Virgin and Child nearly exactly at the center of the panel. On the other hand, it is possible that we are missing even more of the top of the panel. Rather than just 8.5cm, an additional dowel could be missing, and somewhere between 25.7–28.9 cm or so of the panel altogether. In this scenario the panel would have been extremely tall and narrow, improbably so, and the painted composition argues in favor of the former scenario.

Original Frame: Traces of green paint along the raised lip of paint at the edges suggest the decoration of the original frame. In some places two layers of paint are preserved: a lighter green followed by a more translucent green or sometimes brown glaze, suggesting that the green color may have varied, perhaps painted in imitation of a green stone. In a few locations it appears that the green paint on the frame lies over paint from the composition, revealing that the frame was decorated after the painted composition was made. Notably, both the use of a semi-integral frame and the painted decoration are characteristic of Eyckian frames (Verougstraete, 2015, pp. 238–44).

Preparation: The panel was prepared with a white ground, likely containing chalk. Examination with infrared reflectography revealed the presence of an extensive underdrawing executed using a liquid medium (fig. 18).[2] In the underdrawing the artist laid out the most important contours for the Virgin and Child and their drapery, adding only a few hatchmarks to suggest shading in the Virgin’s voluminous robes. He seems to have been more concerned with planning the architecture in the underdrawing, using a combination of incised lines and underdrawing. He incised the arches at the top of the niche and the tracery, while working out the many details of the elaborate architecture in the underdrawing. Presumably there is even more underdrawing present than can be seen in the infrared reflectogram, as the presence of a carbon-containing black paint in the paint layers could be obscuring some underdrawing. The artist added a good deal of shading to the architectural underdrawing, sometimes using rows of short parallel hatchmarks to indicate the shadows cast by narrow columns and at other times long parallel marks across forms to denote large fields of shading. Infrared reflectography did not reveal any underdrawing in the brocade cloth of honor.

The artist made several changes to his underdrawn design when he began painting. He enlarged the Virgin’s facial features slightly and shifted her entire face downwards, as he did with the Child. Otherwise, his alterations to their forms consist of minor corrections to contours. In the architecture the artist made many corrections, shifting the locations of columns and arches.

Paint Layers: The painter created this Eyckian composition in a manner that is consistent with typical Netherlandish painting practice from the mid- to late fifteenth century. The fleshtones were largely created through the application of warm glazes and a selective use of lead white, as can be seen in the x-radiograph. The artist used a dark brown paint to emphasize the contours of his fleshtones. A small number of bright yellow highlights on strands of the Virgin’s hair and the gold trim at her drapery give a convincing sense of naturalism and suggest that light is striking the two figures in a somewhat darkened space, the architectural niche. Similarly, the artist used select highlights to create the sense of light illuminating the luxurious gold-brocaded cloth of honor. The gold threads were created with three colors, a dark brown, dark yellow, and light, but bright yellow, this last applied in a painterly way to create highlights on the gold threads and give the impression of folds in the fabric. The blue cloth now appears mottled, with irregular passages of light blue, but was likely a solid blue originally and likely meant to be silk.[3] The architecture was also painted with a limited palette, primarily consisting of shades of grey, but the painter, often working wet-in-wet, nonetheless achieved a convincing sense of space and subtle shifts of lighting.

The artist introduced a few minor alterations while painting. He initially left a smaller reserve for the white wrapping of the Child, ultimately enlarging the cloth out over the Virgin’s red robe. It appears that the artist first draped the cloth of honor over the edge of the stone step on which the Virgin stands, but then decided to shorten it. This initial position of the cloth of honor would have obscured much of the lettering on the face of the step below, raising the question of whether the inscriptions were planned from the outset. There does not appear to be any underdrawing for the lettering in the infrared reflectogram. The Virgin’s red robe also seems to have draped over the edge slightly.

The painting remains in generally good condition. There are some scattered losses, particularly in the red robe, and some minor abrasion overall, mostly apparent in the warm brown passages. There are some losses near the lower edge of the composition, occasionally extending into the inscription in the pedestal, likely made during the removal of the original frame. The losses are now restored with mimetic retouching, but their locations are easily apparent in the infrared reflectogram and x-radiograph.

Sophie Scully 2023

[1] Initial dendrochronological analysis indicated an earliest possible creation date of 1489 with a more plausible date of 1495 onwards, but this has not been substantiated with additional attempts at dating (complete report housed in the Department of Paintings Conservation). These particular wood boards are exceptionally slow-growing oak, which can impede the accuracy of dendrochronological analysis. Dendrochronological dating of this panel is ongoing.
[2] Infrared reflectography was acquired with an OSIRIS InGaAs near-infrared camera fitted with a 6-element, 150 mm focal length f/5.6–f/45 lens; 900-1700 nm spectral response, April 2022.
[3] Our thanks to Esther van Duijn for offering her opinion on the painting technique in the cloth of honor based on high resolution photographs, by email exchanges, dated April 2022.

Inscription: Inscribed: (on canopy) DOMVS.dEI.EST.ET.PORTA.C[O]ELI ([This is none other than] the house of God and [this is] the gate of heaven [Genesis 28:17].); (on step) IPSA EST [MVLIER] QVAM PR[A]EPARAVIT DOM[INV]S FILIO D[OMI]NI MEI (Let the same be [the woman] whom the Lord hath appointed out for my master's son [Genesis 24:44].)

[Lambert Jean Nieuwenhuys, Brussels, until 1823; said to have come from an old monk in Ghent; sold along with 22 other "tableaux gottiques" on April 23, 1823 for Fr 40,000 to William II]; William II, King of the Netherlands, Brussels (1823–50; sale, de Vries, Roos, and Brondgeest, The Hague, August 12, 1850, no. 3, for fl. 600 to Nieuwenhuys); [Charles J. Nieuwenhuys, London, from 1850]; Alexander James Beresford Hope, London (by 1857–86; sale, Christie's, London, May 15, 1886, no. 30, as by Jan van Eyck, for £315 to Lesser); [Lesser, London, from 1886]; [Sedelmeyer, Paris, 1886]; Henry G. Marquand, New York (1886–89)

Manchester. Art Treasures Palace. "Art Treasures of the United Kingdom," May 5–October 17, 1857, no. 384 (as by Jan van Eyck, lent by A. J. B. Beresford Hope).

London. South Kensington Museum. "Special Exhibition of Works of Art of the Mediaeval, Renaissance, and More Recent Periods," June–October 1862, no. 1051 (as by Jan van Eyck, lent by A. J. B. Beresford Hope).

Leeds Art Gallery. "National Exhibition of Works of Art," 1868, no. 521 (as by Jan van Eyck, lent by A. J. Beresford Hope).

London. Royal Academy of Arts. "Winter Exhibition," 1871, no. 273 (as by Jan van Eyck, lent by A. J. Beresford Hope).

New York. The Metropolitan Museum of Art. "Collection of Dutch and Flemish Paintings by Old Masters, Owned by Mr. Charles Sedelmeyer," Winter 1886–87, no. 30 (as by Jan van Eyck, lent by Henry G. Marquand).

New York. The Metropolitan Museum of Art. "From Van Eyck to Bruegel: Early Netherlandish Painting in The Metropolitan Museum of Art," September 22, 1998–February 21, 1999, no. 48.

Staatsgalerie Stuttgart. "Hans Holbein d. Ä.: Die Graue Passion in ihrer Zeit," November 27, 2010–March 20, 2011, no. 119.

C. J. Nieuwenhuys. Description de la galerie des tableaux des S.M. le roi de Pays–Bas. Brussels, 1843, p. 5, no. 3, catalogues this painting as a work of Jan van Eyck.

W. Burger [Théophile Thoré]. Trésors d'art exposés à Manchester en 1857. Paris, 1857, pp. 152–53 n. 1 [reprinted as "Trésors d'art en Angleterre," Brussels, 1860, with same pagination], discusses it as one of four works attributed to Van Eyck in the catalogue of the exhibition, commenting on the beautiful drapery and excellent architecture but finding the Virgin's head not decisive for the attribution.

[Gustav Friedrich] Waagen. Galleries and Cabinets of Art in Great Britain. London, 1857, p. 190, considers it an early work of Jan van Eyck.

H. G. Hotho. Die Malerschule Hubert's van Eyck nebst deutschen Vorgängern und Zeitgenossen. Vol. 2, Die flandrische Malerei des fünfzehnten Jahrhunderts. Berlin, 1858, pp. 200–201, ascribes it to Jan van Eyck, incorrectly placing it in the collection of the Czar in Saint Petersburg.

W. Bürger [Théophile Thoré]. Musée d'Anvers. Brussels, 1862, p. 25, mentions it in connection with the Van Eyck Madonna in Antwerp ["Virgin and Child by the Fountain," Koninklijk Museum voor Schone Kunsten].

Alfred Michiels. Histoire de la peinture flamande depuis ses débuts jusqu'en 1864. Vol. 2, 2nd ed. Paris, 1866, pp. 318–19, mentions it in his discussion of works by Jan van Eyck.

G. F. Waagen. "Notizen." Zeitschrift für bildende Kunst 3 (1868), p. 127, attributes our picture and a Madonna and Child in the Suermondt collection (now Berlin–Dahlem) to Jan van Eyck and considers ours the earlier of the two.

W. Bürger [Théophile Thoré]. "Nouvelles études sur la Galerie Suermondt à Aix-la-Chapelle." Gazette des beaux-arts, 2nd ser., 1 (1869), p. 11, calls this perhaps a doubtful attribution.

J. A. Crowe and G. B. Cavalcaselle. The Early Flemish Painters. 2nd ed. London, 1872, pp. 114–15, observe that if it really is a work of Jan van Eyck, it is one of his poorest.

Alfred Woltmann. "Die Galerie Suermondt." Zeitschrift für bildende Kunst 9 (1874), p. 195, refers to it in connection with the Suermondt picture as an early work of Jan van Eyck.

William Martin Conway. Early Flemish Artists and their Predecessors on the Lower Rhine. London, 1887, pp. 114–15.

W. Bode. "Alte Kunstwerke in den Sammlungen der Vereinigten Staaten." Zeitschrift für bildende Kunst, n.s., 6, no. 1 (1895), p. 17, calls it probably not autograph.

foreword by [Wilhelm von] Bode. Königliche Museen zu Berlin: Beschreibendes Verzeichnis der Gemälde. 4th ed. Berlin, 1898, p. 100, refers to it as a work probably correctly ascribed to Jan van Eyck.

Ludwig Kaemmerer. Hubert und Jan van Eyck. Bielefeld, 1898, pp. 107–9, fig. 83, ascribes it to Petrus Christus, noting that beyond the differences in handling and conception, the late style of the architecture makes it unlikely that Jan could have made this picture.

Karl Voll. Die Werke des Jan van Eyck. Strasbourg, 1900, p. 127 n. 20, rejects the attribution to Jan, calling it unrefined and weak.

[Hippolyte] Fierens-Gevaert. La Renaissance septentrionale et les premiers maîtres des Flandres. Brussels, 1905, p. 164 n. 2, as dryly painted and perhaps by Petrus Christus.

Salomon Reinach. Répertoire de peintures du moyen age et de la renaissance (1280–1580). Vol. 2, Paris, 1907, p. 360, ill. (engraving).

W. H. James Weale. Hubert and John van Eyck, their Life and Work. London, 1908, pp. 160–61, 202, no. 31 ill. opp. p. 160, rejects the attribution to Jan van Eyck calling it certainly the work of an imitator; notes that the architecture represented could not be dated before the middle of the fifteenth century and dates the picture about 1460.

Emile Durand-Gréville. Hubert et Jean van Eyck. Brussels, 1910, pp. 100–1, ill., 2nd pl. after p. 100, considers it the finest and most faithful copy of an [apparently lost] original which he ascribes to Hubert van Eyck; based on a photograph, tentatively ascribes it to Petrus Christus.

E. Durand-Gréville. "Les deux Petrus Christus." Revue de l'art ancien et moderne 30 (July–December 1911), p. 141–42, 203, 206, ill., based on reproductions, ascribes it to the father of Petrus Christus, "Petrus Christus le Vieux," and dates it tentatively about 1415–20; believes it is based on a lost original by Hubert van Eyck.

Morton H. Bernath. New York und Boston. Leipzig, 1912, p. 52, as by Christus.

W. H. James Weale and Maurice W. Brockwell. The Van Eycks and their Art. London, 1912, pp. 174–75, 207, no. 39, call it a "mere imitation" and date it probably about 1460; mention a "copy or imitation" of this picture in the collection of David Sellar [now Museum of Fine Arts, Budapest, as Petrus Christus].

Walter Cohen in Allgemeines Lexikon der bildenden Künstler. Ed. Ulrich Thieme. Vol. 8, Leipzig, 1913, pp. 125–26, lists it as a work of Petrus Christus.

Martin Conway. The Van Eycks and Their Followers. London, 1921, p. 69, observes that "it may have been Petrus Chistus who so carefully copied this Madonna group" after Jan's "Virgin and Child" in Antwerp.

Max J. Friedländer. Die altniederländische Malerei. Vol. 1, Die Van Eyck, Petrus Christus. Berlin, 1924, p. 164, places this picture between Jan and Christus in style and quality.

L. Maeterlinck. Une école préeyckienne inconnue. Paris, 1925, p. 28, fig. 58, observes that the manifest inferiority of the picture does not permit an attribution to either Jan or Hubert.

Malcolm Vaughan. "Paintings by Petrus Christus in America (part 1)." International Studio 89 (January 1928), ill. p. 31, as by Christus.

G. H. Edgell. Letter to Harry B. Wehle. September 26, 1933, [Edgell was Dean of Harvard University School of Architecture] guesses a date between 1460 and 1470 on the basis of the architecture.

Harry B. Wehle and Margaretta Salinger. The Metropolitan Museum of Art: A Catalogue of Early Flemish, Dutch and German Paintings. New York, 1947, pp. 20–23, ill., attribute it to Petrus Christus, perhaps copying a work of Jan van Eyck; record and translate the inscription and identify the statuettes in the architecture.

Ludwig Baldass. Jan van Eyck. New York, 1952, p. 275 n. 2, attributes it to Christus, calling it a free rendering of Jan's "Virgin and Child by the Fountain".

Germain Bazin. "Petrus Christus et les rapports entre l'Italie et la Flandre au milieu du XVe siècle." Revue des arts 4 (December 1952), p. 199, mentions it as a painting recently added to Christus' oeuvre.

Erwin Panofsky. Early Netherlandish Painting: Its Origins and Character. Cambridge, Mass., 1953, vol. 1, pp. 350, 416 n. 5, calls it "Madonna in the Aedicula," a copy of about 1460 after Jan's "Virgin and Child by the Fountain;" reads DO[MI]N[V]S, not DOM[U]S in the footpace inscription [see Ref. Wehle and Salinger 1947].

Josephine L. Allen and Elizabeth E. Gardner. A Concise Catalogue of the European Paintings in The Metropolitan Museum of Art. New York, 1954, p. 18.

Josua Bruyn. Van Eyck problemen. Utrecht, 1957, pp. 43, 104 n. 2, pp. 123–28, 132, 134, 150, fig. 56, considers it the work of an unknown pupil of Jan van Eyck, possibly the master responsible for the Prado "Fountain of Life," noting that our Virgin's cloth of honor has the same pattern as the cloth behind John the Evangelist in the Prado picture; observes that the architecture resembles that in a drawing of the Annunciation at Wolfenbüttel (fig. 57).

Mirella Levi d'Ancona. The Iconography of the Immaculate Conception in the Middle Ages and Early Rennaissance. [New York], 1957, p. 35 n. 86, p. 70 n. 162.

Pierre Quarré. "Fragment d'un primitif de la Chartreuse de Champmol." Revue des Arts 7 (March–April 1957), pp. 62–63, ill., publishes a fragment (now in Musée des Arts Décoratifs, Paris) of an Eyckian panel which hung in the Chartreuse de Champmol, on which are depicted portions of an aedicula similar to the one in our picture; states that these architectural elements originally framed a Madonna and Child which must have been comparable to the figures in our panel.

Erik Larsen. Les primitifs flamands au Musée Metropolitain de New York. Utrecht, 1960, pp. 38–39, 109–10, fig. 4, believes it is a sketch by Jan van Eyck which was abandoned and later finished by Christus.

R. H. Wilenski. Flemish Painters, 1430–1830. New York, 1960, vol. 1, pp. 12, 30, 73; vol. 2, pl. 125.

Raymond Bouyer Giorgio T. Faggin in L'opera completa dei Van Eyck. Milan, 1968, p. 97, ill., disusses it as a copy of the Antwerp panel (no. 29) which may date from about 1450.

Lola B. Malkis Gellman. "Petrus Christus." PhD diss., Johns Hopkins University, Baltimore, 1970, pp. 497–99, no. 30, fig. 81, rejects the attribution to Christus.

Charles Sterling. Letter. February 20, 1971, says that he would call it "Follower of Jan van Eyck close to Petrus Christus," and adds that the architecture depends on Rogier's Granada–Miraflores altarpiece, probably painted about 1438–40.

Joel M. Upton. "Petrus Christus." PhD diss., Bryn Mawr College, Bryn Mawr, Pa., 1972, pp. 420–21, no. 44, rejects the attribution to Christus and finds it close in style to the dry rendering of the Frick Madonna, also underservedly attributed to Christus.

Lino Moretti. G. B. Cavalcaselle: disegni da antichi maestri. Exh. cat., Fondazione Giorgio Cini, Venice. [Vicenza], 1973, p. 73, ascribes our panel to an imitator of Jan van Eyck; publishes the sketch Cavalcaselle made after it at the 1857 Manchester exhibition and notes that he queried its attribution to Jan in the margin of his catalogue.

Susan Urbach. "Domus Dei Est et Porta Coeli: Megjegyzések Petrus Christus Madonna-Képének Ikonográfiájához." Építésépítészettudomány 3–4 (1974), pp. 349–51, ill.

Francis Haskell. Rediscoveries in Art: Some Aspects of Taste, Fashion and Collecting in England and France. Ithaca, N.Y., 1976, p. 69, pl. 160.

David Robertson. Sir Charles Eastlake and the Victorian Art World. Princeton, 1978, p. 185, refers to it as by Petrus Christus.

J. W. Niemeijer. "A Conversation Piece by Aert Schouman and the Founders of the Hope Collection." Apollo, n.s., 108 (September 1978), p. 187.

Elisabeth Dhanens. Hubert and Jan van Eyck. New York, 1980, p. 365, ill., with the works of "Anonymous Epigones and Imitators".

Larry Silver. "Fountain and Source: A Rediscovered Eyckian Icon." Pantheon 41 (April–May–June 1983), pp. 95–96, 102 n. 4, ill., calls it "After Jan van Eyck" and dates it mid–fifteenth century.

John Pope-Hennessy. "Roger Fry and The Metropolitan Museum of Art." Oxford, China, and Italy: Writings in Honour of Sir Harold Acton on his Eightieth Birthday. Ed. Edward Chaney and Neil Ritchie. London, 1984, p. 231.

Erik Hinterding and Femy Horsch. "Reconstruction of the Collection of Old Master Paintings of King Willem II." Simiolus 19, no.1–2 (1989), p. 56, as style of Jan van Eyck.

Erik Hinterding and Femy Horsch. "'A small but choice collection': The Art Gallery of King Willem II of the Netherlands (1792–1849)." Simiolus 19, no.1–2 (1989), p. 10 n. 19.

Peter Klein. Letter to Maryan Ainsworth. April 15, 1991, dates the panel from 1503 upward through dendrochronological analysis.

Craig Harbison. Jan van Eyck: The Play of Realism. London, 1991, p. 163, dates it to the mid-15th century and calls it one of at least eight very exact copies of van Eyck's Antwerp Virgin and Child; suggests that with such careful copying Netherlandish artists were—in the manner of earlier artists who laboriously copied Byzantine icons—attempting to preserve the "magic formula" of van Eyck's "icons".

Jos Koldeweij. "Book Reviews: Maryan Ainsworth with contributions by Maximiliaan P. J. Martens, 'Petrus Christus: Renaissance Master of Bruges'." Simiolus 23, no. 4 (1995), pp. 269, 271–72, finds it odd that Ainsworth does not discuss this picture.

Katharine Baetjer. European Paintings in The Metropolitan Museum of Art by Artists Born Before 1865: A Summary Catalogue. New York, 1995, p. 244, ill.

Christian Heck. Letter to Maryan Ainsworth. April 15, 1996, finds this panel a good example of the identification in western medieval art of the Virgin with Jacob's ladder, as shown by the upper inscription.

Maryan W. Ainsworth. "A Meeting of Sacred and Secular Worlds." From Van Eyck to Bruegel: Early Netherlandish Painting in The Metropolitan Museum of Art. Ed. Maryan W. Ainsworth and Keith Christiansen. Exh. cat., The Metropolitan Museum of Art. New York, 1998, pp. 65, 81, 211, 220, 222–23, no. 48, ill. (color), as "Netherlandish" about 1500.

Maryan W. Ainsworth. Gerard David: Purity of Vision in an Age of Transition. New York, 1998, p. 263, fig. 247.

Elizabeth A. Pergam. "From Manchester to Manhattan: The Transatlantic Art Trade After 1857." Bulletin of the John Rylands Library 87, no. 2 (2005), pp. 66–67, 86, 88, reproduces a sketch by George Scharf recording the placement of this work at the Manchester exhibition of 1857.

Jenny Graham. Inventing Van Eyck: The Remaking of an Artist for the Modern Age. Oxford, 2007, pp. 151, 227 n. 97, fig. 55.

Elsbeth Wiemann in Hans Holbein d. Ä.: Die Graue Passion in ihrer Zeit. Ed. Elsbeth Wiemann. Exh. cat., Staatsgalerie Stuttgart. Stuttgart, 2010, pp. 366–69, no. 119, ill. (color).

Susan Urbach. Early Netherlandish Paintings. London, 2015, vol. 1, pp. 75, 77 n. 28, under no. 4.

Larry Silver. "Eyckian Icons and Copies." Making Copies in European Art 1400–1600: Shifting Tastes, Modes of Transmission, and Changing Contexts. Ed. Maddalena Bellavitis. Leiden, 2018, p. 137, fig. 3.4 (color).