RKD STUDIES

Oil on canvas, 76.5 x 71 cm¹
Antwerp, The Phoebus Foundation

This painting briefly attracted a great deal of attention, due to having been a surprise discovery. It was sold at a provincial auction in England for three pounds, then sold at Christie’s in London for 2.800 guineas. At the 1937 Hals exhibition, it was listed as no. 8 in the catalogue. Together with the large group of other depictions of fisherchildren, its authenticity was fundamentally questioned by Maurits Michel van Dantzig, Numa Trivas, and most other critics of the attributions to Hals at the time.² Slive concurred: ‘To judge from reproductions, it is a nineteenth-century painting done in Hals’ style’.³

In contrast to my own judgment of 1972, I have come to the opinion that this painting is the most artistically outstanding example from the group of popular half-length figures from the Hals workshop, after having inspected it first-hand before and after the 2016 cleaning by Martin Bijl.⁴ There are several detail-areas that are painted in a strikingly fresh manner – especially the heads of the two boys. Accuracy of modelling is combined with a clear brushstroke that only Hals was capable of. The soft and confidently color-graded faces, and the hat of the boy in the foreground, that is elegantly modelled with stripy brushwork, indicate a handwriting that is as routine as it is masterly. It can be traced as far as the boy’s shoulder area. Conversely, the remaining areas – of the sleeve, hand, jacket, and basket – were painted by a coarser hand. This difference in ‘hands’ became even more apparent after the cleaning in 2016. The picture then lost its shellac-like varnish, revealing a play of delightful color nuances especially in the faces. The softness in the modelling and the creation of sculptural details in the faces are in accordance with the style of the commissioned portraits by Hals, created between 1634 and 1637. The pointed manner of using highlights and shadows also fits into this period, neither earlier nor later.

In 2019, Anna Tummers, Arie Wallert and Nouchka de Keyser published the results of a technical investigation commissioned by Christie's.⁵ The occasion was the claim by Han van Meegeren's son Jacques that the present painting was a forgery made by his father. Already in 1937, restorer Van Dantzig had described the painting as a late 19th-century forgery and still by 2007, Eddy de Jongh classified it as such.⁶ The scope of the 2019 study included also Malle Babbe from the Rijksmuseum, forged by Van Meegeren (X5), and the Fisherboy in a landscape from the Royal Museum of Fine Arts in Antwerp (A4.2.20). The determination of the lead white pigments was particularly clear: ‘[…] the sample from Malle Babbe shows entirely different isotope ratios […], indicating that the lead ores in the lead white used by Van Meegeren came from a completely different location than Hals's lead white. These lead isotopes ratios are not consistent with materials used by seventeenth century painters’.⁷ The evaluation of the further findings made it clear that the present painting belongs to the 17th century and originates from the Hals workshop. It also suggested a dating of c. 1634-1637 and the signature could be determined as forming an original component of the paint layer.

The diagonally rising dune landscape in the present panting’s background conforms to a pattern that can be found in several other genre paintings by the Hals workshop. In its cool tonality and draughtsman-like brushwork, the landscape is recognizable as a contribution by Pieter de Molijn (1595-1661), whose hand is visible in the background of several other works by Hals. The rolling dune landscape with small figures in the distance and the ochre and green tones are related to Molijn’s Landscape with a cottage, dated 1629 [1].

The boy with the shadowed face in the foreground reappears as – or at least strongly resembles – the model in Young boy in profile in Washington (A3.31). That painting was also executed by Hals in the facial area, while the remainder must be considered to be a work by an artist from his immediate circle. The correspondence between both representations suggests a similar time of execution.

The motif of the crab on the boy’s finger, or of children playing with crabs in general, is recurrent in artworks of the period. It first appeared in the work of Italian artists, for example in a drawing and painting by Sofonisba Anguissola (c. 1535-1625), both undated.⁸ Similar subject matter can be found in a half-length figure by Caravaggio (1571-1610), depicting a young man being bitten by a lizard.⁹ Also worth mentioning is Annibale Carracci’s (1560-1609) composition of children teasing a cat with a small crab, dating from around 1590.¹⁰ But the subject matter of pain inflicted through the bite of a crab can be found in Holland as well, for example in the picture Boy with fish still-life by Petrus Staverenus (c. 1610/1612- after 1654), who was heavily influenced by Hals and worked in Haarlem and The Hague.¹¹ All these representations certainly refer to worldly wisdom, be it about the pain, the children’s courage, or the multiple symbolisms of the crab.

A3.30
© The Phoebus Foundation, Antwerpen

1
Pieter de Molijn
Landscape with a cottage, dated 1629
New York City, The Metropolitan Museum of Art, inv./cat.nr. 95.7

Oil on panel, ø 19 cm
Washington, National Gallery of Art, inv.no. 2009.113.1

In contrast to Slive, I have no doubts about dating this painting to the 17th century, and even less about attributing it to Frans Hals and his workshop.¹² The National Gallery of Art assigns the roundel to Judith Leyster (1609-1660), in accordance with the earlier attribution from the art trade and Hofrichter.¹³

This handsome facial study either depicts the same or a similar, slightly older, model as is depicted in the foreground of the Two fisherboys (A3.30). The lighting and the modelling are different, though, as is the incline of the head. Thus, we are looking at two separate studies, executed independently from each other but probably during the same period. As it often happened when the facial features of a figure point to a certain master, but the rest varies stylistically, this painting was also attributed to the circle of pupils of Hals. However, two distinct manners of execution are discernible in this case. The facial contours were drawn in a very assured manner with a few soft lines of the brush. This area, including the ear, displays a painterly talent which cannot be identified anywhere in the work of Hals’s pupils, and neither in the secure works by Leyster. This is Hals’s own handwriting, which contrasts in the present painting with the precise contours and smooth completion of the shoulder and collar. The peculiarly dashed brushwork in the hair is striking, but also does not match the style of Leyster. Neither does the precisely modelled collar. Therefore, the painting must have been either designed by Frans Hals with the face executed by him, or it was one of his unfinished studies which was completed to form a portrait-like representation by a different hand.¹⁴ I am pleased that Hofrichter agreed to my attribution in a recent exchange of letters.¹⁵

A3.31

Oil on canvas, 111.1 x 86.7 cm
Sarasota, The John and Mable Ringling Museum of Art, inv.no. SN251
Pendant to A3.33

The larger size of the pendant Portrait of Maritge Vooght Claesdr. (A3.33) and the cut-off coat of arms indicate that the present painting was trimmed on all sides. Both paintings are executed on canvas of identical thread count, probably woven on the same loom.¹⁶

Hals had already painted Pieter Jacobsz. Olycan (1572-1658) several years earlier (A3.25). The presence of two pairs of portraits of him and his wife is confirmed by the 1666 inventory of Olycan’s daughter Geertruyt (1603-1666), that records one pair in the entrance hall and another elsewhere in the house.¹⁷ Pieter Jacobsz. Olycan was the politically most active and most influential member of his clan. He had been educated in Rome and had undertaken business travels to Spain and the Baltic countries. He owned several breweries and 47 pieces of land in Haarlem. He repeatedly held offices as councilor, lay judge and mayor of the town of Haarlem, which he represented at the States General in 1631 and 1640. His powerful appearance was captured in Hals's individual portraits as well as in the group portrait of the Calivermen civic guard of 1630, which shows him as colonel.¹⁸ In Hals's 1639 portrait he is wearing a dignified fur-trimmed coat.

Upon close examination of this painting, two different hands are evident in the execution: Olycan's cuffs and his bulky hands with gloves were not painted by Hals himself. In addition, changes were made in the hair on the head, especially above the ear. The coat with the fur seam was also painted in a coarser hand. In contrast, the truncated coat of arms shows the same cool and garish colors as in the coats of arms in the companion piece as well as those in the two Olycan portraits of 1625 (A1.17, A1.18), which have now been rendered invisible by a neutral covering, since their chemical analysis had revealed them to be later additions. In spite of the different states of preservation and the unresolved reworkings, a precise comparison of the head areas in both portraits of Pieter Jacobsz. Olycan (A3.25, A3.32) is instructive with regard to our understanding of the sitter’s personality as well as Hals’s stylistic development.¹⁹ Olycan’s spontaneous facial expression is especially marked in the present portrait, which also relies on the clarity of Hals’s brushwork in this central area.

Oil on canvas, 126.4 x 93.2 cm, inscribed and dated upper left: AETAT SVAE 62/AN° 1639
Amsterdam, Rijksmuseum, inv.no. SK-C 139

This portrait of a distinguished woman with an alert gaze is overshadowed by its pendant, the Portrait of Pieter Jacobsz. Olycan (A3.32) in his wide black coat. It is preserved it its original composition, apart from the coat of arms that was strengthened later in thick paint by another hand, and the widened cover of the precious family bible that is out of line with the perspective. The seated position repeats that of the sitter’s sister, Cornelia Claesdr. Vooght (* c. 1578) (A3.20), but Maritge has a slightly lower chair, so that the erect figure contrasts markedly with the background.

At first glance, the paint surface appears consistent and – in comparison to the pendant – perfectly preserved. However, on close inspection, smoothing overpainting is recognizable in many parts. Hals’s angular brushwork is only preserved unchanged in a few areas: on the white bonnet on the right, within the face only in the area of the lips, on the lower edge of the nose, the shadow of the nose and on the left fold of the cheek [2], on the black silk parts in both sleeves and on the cuff of the left arm. Smoothing overpainting appears on the left side of the bonnet, on the millstone collar including the cast shadow from the face, the forehead, the area of eyelids and cheeks, the shadow on the right temple and the right cheek, the left hand and the entire right hand, parts of the adjoining white cuff as well as the fur trim and the folds of the black coat. The effect is probably that of a lighter and more static representation of the sitter.

The shared fate of the ambitious portraits of the Vooght sisters and their husbands (A3.19 and A3.20, A3.25, A3.32 and A3.33) is striking. All five pictures were reworked extensively, some even shortly after their creation. In all cases the surface of faces and hands was smoothed, but also the clothing was executed more coarsely. In the 1631 Portrait of Cornelia Claesdr. Vooght (A3.20) it is not clear whether the presentation of the fur and the dress fabric were part of the initial execution or of the reworking. However, in the present Portrait of Maritge Claesdr. Vooght of 1639, there is an area in the lower left where a differently structured paint surface shines through. These observations allow the conclusion that the patrons commissioned changes, be it during the first stage of execution or subsequently, which were carried out by another hand than that of Frans Hals himself.

The dissatisfaction of prominent patrons with Hals’s style of painting, which arises from the above observations, was recently associated with Joachim van Sandrart’s (1606-1688) report about the successful young painter Jacob Adriaensz. Backer (1608-1651) in Amsterdam. According to this, Backer ‘was famous above other due to his speed / as he showed me himself a woman/ who / to have herself painted / had arrived from Haarlem / and travelled back home on the same day / which he had in such a short time with the face / collar / fur / coat / as well as other clothing and both hands / turned into a life-size half figure good-looking and well-made’.²⁰ This passage is quoted in the Wikipedia entry on Jacob Backer with reference to the large-scale Portrait of a lady in the Metropolitan Museum of Art, signed Rembrandt f. and dated 1635.²¹ It is commented as follows: ‘perhaps this portrait of a woman in an armchair by Backer seemed more attractive to the sitter mentioned by Sandrart than the portraits by Frans Hals of her Haarlem contemporaries, the Vooght sisters, and that is why she travelled to Amsterdam for her portrait to be done’.²²

Regardless of the still open attribution of the New York portrait, the question arises against which contemporary alternatives Hals’s female portraits of the 1630s would be measured in Haarlem. Which painters were available there apart from Hals, what did their work cost and how busy were they? An associated question concerns the overpainter, who smoothed faces and hands, and the assistant – or could it have been the same person? – who executed the folds of the dress and the fur. When comparing these interventions with works by potential competitors – Salomon de Bray (1597-1664), Johannes Verspronck (c. 1600/1603-1662), Pieter Soutman (c. 1593/1601-1657) – a participation by Verspronck seems to me only plausible in the repetition of the Portrait of Maritge Claesdr. Vooght (A3.33a), as well as possibly for the reworking of the head in the Portrait of a man in Berlin (A3.18). The latter could have been carried out within the Hals workshop even before Verspronck set out independently. Stylistically different, yet close to Hals’s style are the abovementioned areas in the portraits of Nicolaes van der Meer (A3.19), Pieter Jacobsz. Olycan (A3.25, A3.32) and the present portrait. Accordingly, it would have been Hals himself who delegated the reworking to assistants.

2
detail of cat.no. A3.33

Oil on panel, 68.5 x 57.8 cm
Sale New York (Sotheby’s), 6 June 2012, lot 36
Pendant to A3.25 [3]

This is a very well painted partial copy after the 1639 Portrait of Maritge Claesdr. Vooght (A3.33), reduced to a bust-length portrait. It is considered to be the pendant of the Portrait of Pieter Jacobsz. Olycan (A3.25), yet the pair was separated at the sale of the Watney collection in 1967.²³ It is probably also identical with one of the portraits listed in the 1666 inventory of Olycan’s daughter Geertruyt.²⁴

The copy’s style is less reminiscent of Hals and his workshop and more of Johannes Cornelisz. Verspronck (1600/1603-1662), as Biesboer already observed.²⁵ It is conceivable that the commission process was similar to that of the reworking the portraits of the sitter’s brother-in-law and sister (A3.19, A3.20), and of the copy after Hals’s 1650 Portrait of Nicolaas Stenius (A3.54) by Ludolf de Jongh (1616-1679), dated 1652 (A3.54a), as a copy by a different workshop may have been obtained cheaper, faster or more reliably.

The present painting is of particular interest from the point of view of technique and the largely preserved coloring. In today’s appearance it can be compared directly to Hals’s – reworked – model and gives a good impression of the thinly painted half-tones, especially in the shaded parts of the face – which appear lighter than in the model.

Oil on canvas, 80 x 66.5 cm, monogrammed right of center: FH²⁶
St. Petersburg, State Hermitage Museum, inv.no. ГЭ-982

Narrow canvas strips were added to the painting's top edge and on both outer sides. Stylistically, it falls into the period between Hals's warm-toned, softly modelling phase of the 1630s and the more planar late style that is dominated by shades of grey. The composition is unquestionably by Hals, and he surely executed the face himself. The accurately captured turn of the gaze and the movement in the facial features can only be attributed to the master himself. Hals's typical brushwork is recognizable in the execution of the hair on the shaded side, and in the highlights on the shimmering parts of the illuminated side. However, the contours around the nose, eyes and mouth have been smoothed over, and the line between the nostril and the cheek was also flattened. The tips of the hair above the eyebrows and along the cheek seem to have been cut off, and overpainting has made the transition from the cheek to the hair seem unnaturally smooth. Only a detailed scientific analysis would be able to clarify how much the area of the face and the collar was affected by later reworking. Disregarding any later overpainting, the primary depiction is a cooperation between Hals and an assistant. The design of the hands lacks accuracy, both in the flat upper hand that vanishes into the dark and in the lower semi-dark hand which seems stunted. These defects are not the result of a smoothing overpainting but areas of careless execution, in other words the work of a helper.

Around 1670, the English engraver Edward Davis (c. 1640-c. 1684) created a copper engraving (C20) which combined the present portrait with Hals's Leipzig Peeckelhaering (A1.51).

A3.34
Image is used from www.hermitagemuseum.org, courtesy of The State Hermitage Museum , St. Petersburg, Russia

Oil on canvas, 84.2 x 67.6 cm, inscribed, dated, and monogrammed upper right: AETĀ. SVAE 56 / AN 1640 / F H²⁷
Haarlem, Frans Hals Museum, inv.no. OS 2008-91

The immediacy of the facial expression, the tonality of the color and the brushwork which is especially noticeable in the area of the head, all associate this painting with Frans Hals. At the same time, the remaining areas of the portrait appear flat and lack the creative accents that typically model Hals's body shapes. The figure, the dark coat and the cap of the seated gentleman are not executed in detail, even taking into account any darkening of the thin paint layer. The back of the chair is also merely sketched, and the collar is only indicated in its outline. The hand is treated summarily, and its two-dimensionality falls behind Hals’s hands and fingers which are always anatomically understood and fleshly painted. All above-mentioned passages do not have the hallmarks of Hals’s involvement in their execution. We must therefore assume that Hals designed the composition but limited his contribution to the first sketch of the sitter’s face – which stands out as an island of brightness in the otherwise generally dark and flat painting. Hals uses exact observation of the moving eyes, eyelids, and eyebrows to capture temperament and typical character. This psychological constellation is represented in the sketch of the face. But the painterly execution in the details of the facial features differs from Hals’s rhythmically sweeping and painterly blended brushwork, especially in the dashed contours of the ear, beard, nose, mouth, and the contours of the eyelid and creases around the eyes. This execution is most likely by an assistant, probably Jan Hals (c. 1620-c. 1654). The areas of the collar and hand correspond to his rigid draughtsman-like style than can also be observed in his portraits of 1644 and 1648 (A4.3.13, A4.3.25).

In the catalogue of the 2013 Haarlem exhibition, Hillegers proposed to identify the sitter as Willem Warmont (1583/1584-1650).²⁸ Indeed, the age of the Haarlem brewer and mayor matches the man’s age. Nevertheless, to me the facial features do not seem to support this identification. The more voluminous nose and the pointier nose profile, as well as the wide and prominent upper eyelids set the present painting’s sitter apart from Van Warmont’s confirmed likeness in the Banquet of the officers of the Calivermen civic guard of 1627 (A2.8A) [4].

In 2014 Frans Grijzenhout discovered an inventory of the possessions of Nicolaes Noppen (1600/1605-1657) and his wife Geertruijt Gerrits. van Santen (1605/1610-1675), drawn up in 1676. Aside from portraits of the husband and wife (A1.99, A1.100) the inventory lists a portrait by Frans Hals of the father of Geertruijt, Gerrit Jansz. van Santen (1583/1584-c. 1653), which Grijzenhout tentatively matches to the present painting.²⁹ Gerrit Jansz. van Santen was a distiller, and also a Mennonite preacher, which would be in keeping with the present sitter’s unembellished clothing.

A3.35

4
detail of cat.no. A2.8A, head of Willem Warmont
Frans Hals (I)
Banquet of the officers of the Calivermen civic guard, 1627
Haarlem, Frans Hals Museum

Oil on canvas, 85.2 x 68.1 cm, inscribed and dated left of center: AETAT SVAE 53/ AN° 1640
Ghent, Musée des Beaux-Arts, inv.no. 1898-B
Pendant to A1.72 [5]

The painting is one of the latest representations with the sitter depicted inside a painted oval frame. We do not know if this composition was decided on by the painter or, more likely, specified by the patrons. This portrait of a very resolute-looking woman was offered at auction in Amsterdam in 1797 and in Paris in 1876, together with the 1635 male portrait (A1.72), likewise in a feigned oval.³⁰ Afterwards, the two pictures went separate ways. It is unknown if and how they belonged together, as pendants or as parts of a family gallery.

The central part of the woman’s face shows a spirited expression and a spontaneous and convincingly captured turn towards the viewer, which only ever appears in Hals’s own achievements. His characteristic brushwork is present in the areas of the eyes, nose, and mouth [6]. But at the same time, there are somewhat finicky brushstrokes in the facial plane and on the hands that were done by another hand. Individual skin creases, especially the shadow line decreasing from the nose to the right corner of the mouth, are executed hesitantly and lack the master’s visual subtlety. There is no angular, rhythmically dotted pattern of lines in a diagonal direction that is typical for Frans Hals. The hairline, bonnet, cuffs, and collar are treated schematically and hesitantly. Overall, Hals is likely to have designed the composition and corrected it after the execution, which was left to an assistant. In addition, later interventions disturb the overall impression, such as the pink contour on the left cheek, the filled in nostril and the cast shadow under the nose. Stylistically, there is a relationship to the female portrait that was sold at Sotheby’s New York in 2008 (B15).

6
Detail of cat.no. A3.36
Photo: Hugo Maertens

Oil on canvas, 65.7 x 48.7 cm, monogrammed lower right: FH
Sale Paris (Christie’s), 24 February 2009, lot 73

The position of the man in this portrait is turned to the right and can be assumed to relate to a lost pendant. Based on the measurements and the size of the head, the only candidate among the preserved paintings would be the Portrait of a woman in London (A1.98) which, however, has a seated position that would not be easily compatible with the present painting. In addition, the coloring of that picture in its mostly brown shades differs from the cooler tonality modelling the present portrait.

The sitter’s characterful face with the prominent lump on the forehead is a typical performance by Frans Hals, executed as far as the short reddish brushstrokes on the cheek and forehead. This immediacy of observation and clarity of representation is lacking in the other areas of the painting. The collar with its hesitantly executed lace is as flat as the area of the arm with its grey stripes running down in parallel. The hand with the book is also incongruous and anatomically too small. The painting, which is so lively in its main area, was executed in other parts by a different hand

A3.37
© Christie's, Paris

Oil on canvas, 117 x 91.5 cm
Glasgow, The Burrell Collection, inv.no. 35.276

In its affected gestures and balancing posture, this portrait of a somewhat narrow-chested gentleman, diverges from Hals’s static figures. The execution of the body, hands, gloves, and collar follows Hals’s style without achieving his clear plasticity. It must be the case that these areas were executed by an assistant inspired by Hals’s characteristics of painting. In the present condition, later overpainting is also evident.

On 16 March 1937 the painting was put up for sale by a certain Pamela Farmer, heiress of the estate of Nonsuch. At the auction the painting was bought by art dealer Arthur de Casseres for £12,705. The sensational result of the sale was reported internationally and was described as one of the most important art finds for years – an art dealer with a reputation for making big ‘discoveries’, this being his third in just two years. Between 16th March and 30th July 1937, the portrait passed into the hands of D. Katz, a Dutch art dealer. In 1948 William Burrell (1861-1958) bought the work from the art dealer Frank Partridge and Sons for the substantial sum of £14,500, meaning that the painting passed from Katz’s hands to Partridge’s at some point between 1937 and 1948. A stamp on the stretcher also notes that the restorers De Wild worked on the painting before Burrell bought it.³¹

In the early 20th century this painting was regarded as a genuine Hals.³² However, Slive omitted the work from his oeuvre catalogue of 1970-1974, not even mentioning the work in the section with rejected works. Correspondence between Slive and then curator of the Burrell Collection in 1982 shows that Slive was ‘disturbed by virtually everything from the model’s head down to the bottom of the picture’. He complained, specifically, of the proper right shoulder appearing not to fit the subject’s neck, the silhouette on the right side of the painting appearing dull and awkward, and the folds in the drapery seeming, again, awkward.³³ Most of the details criticized by Slive have since become explicable through workshop practices and the history of the picture. Weaknesses in the modelling of the clothes, especially the folds of the dark suit, suggest the hand of an assistant involved in the execution, possibly Jan Hals (c. 1620-c. 1654). Nevertheless, these observations are secondary to the expressive face that seems to turn spontaneously towards the viewer. The head, which is modelled with rich contrasts throughout, the lively eyes, the jauntily added moustache and the movement conveyed in the mouth unequivocally demonstrate Hals’s generous and confident style of painting [7]. He probably sketched the entire figure and then took on the modelling of the head himself. Overpainting around the head most likely contributed to obscuring the quality of this area and preventing an attribution to Frans Hals. The underlying contour of a wide-brimmed hat is discernible, now covered by the background color, as are some of the edges of the white collar. Even the lowered contour of the shoulder that Slive criticized can be regarded as the result of later reworking. An original higher shape is recognizable in the picture’s appearance today. It is to be hoped that these deforming interventions, which obscure the composition, will be reversed in the intended restoration. As in all comparable cases, it would certainly be worthwhile to reveal the original composition [8].

A3.38
© CSG CIC Glasgow Museums Collection

7
Detail of cat.no. A3.38
© CSG CIC Glasgow Museums Collection

8
Claus Grimm
Reconstruction of cat.no. A3.38

Oil on canvas, 61.6 x 51.4 cm
Cambridge (Massachusetts), Fogg Art Museum, inv.no. 1930.186

The painting was cut on all sides. An overly harsh cleaning treatment has almost bleached the bridge of the sitter’s nose towards the forehead to white, which is a blemish. The handling of the collar, the beard and the hair on the head is uncharacteristically smooth for Hals. Nevertheless, the execution of the central area of the face with the sharply observing eyes, positioned with a few precise lines, should be autograph. The clarity of the characterization, combined with the simultaneous confident capture of different levels of brightness, lifts this area above and beyond comparable tasks as performed by the workshop. The folded cloth that serves as a background is of interest. It may indicate an original format that was considerably larger. It is not clear whether the sitter was a priest, as the black cap was also worn by others, especially older gentlemen (for instance A4.3.54). Nevertheless, the frontal position indicates a person of particular dignity. As published by Slive, the picture’s old stretcher bears the seal of the Polish king Stanislaw Augustus, in whose collection it was in the late 18th century.³⁴

A3.39
Photo © President and Fellows of Harvard College