Gilding, Alteration and the Question of Originality in Whistler’s Frames

The online catalogue raisonné of James McNeill Whistler’s oil paintings, edited by Margaret F. MacDonald and Grischka Petri, is unusual in its inclusion of all that is known – up to 2020 – about the artist’s frames.1 It evaluates their originality in relation to each painting and summarises Whistler’s statements on gold, both in relation to frames and in the titles he gave to his paintings.2 The catalogue includes research by Sarah Parkerson Day, who documented frames in many collections, defined types and terminology for the frames that Whistler specified in his lifetime and provided the framing histories of the paintings as known in 2018.3 Day’s study established an essential framework for documentation and formal classification of Whistler’s frames; however, it did not attempt a sustained technical account of the gilding and painted ornament on individual frames.

This article begins to address that gap by examining six Whistler frames in the Tate collection, integrating focused technical examination; microscopy; cross-sections from paint and gilding; and pigment analysis with archival correspondence and object histories. It begins by discussing two painted frames, those around Whistler’s Nocturne: Blue and Gold – Old Battersea Bridge 1872–5 (YMSM 140; Tate N01959) and Nocturne: Blue and Silver – Chelsea 1871 (YMSM 103; Tate T01571).4 It considers a modern replica of the folding screen Blue and Silver: Screen, with Old Battersea Bridge 1871–3 (YMSM 139; The Hunterian, University of Glasgow), made for the 2026 exhibition James McNeill Whistler at Tate Britain, London.5 Featuring the artist’s largest landscape painting, the screen is dated to the same period as the two painted frames and provides insight into how they were made. The article goes on to examine four frames often referred to as ‘Grau frames’. Around the date of an exhibition of his work at Goupil Gallery, London, in 1892, Whistler made a significant shift away from his earlier frame aesthetic to establish a final frame design for posterity. This design was named after the maker Frederick Henry Grau (1859–1892). Unlike the painted frames, three of the Grau frames experienced drastic alterations after accession.6 This raises questions as to whether the assumption that the frames had been painted by Whistler increased their status and saved them from such radical alteration, while objects constructed by others to his design were more likely to have been altered.

Close examination and an understanding of each object’s history were brought together with research into the artist’s intentions to inform treatments in advance of the 2026 exhibition. Historic treatments and alterations to Whistler’s frames were seldom recorded until recent years, with a trail of interventions and rationales often difficult to establish before about 1980. A review of the two painted frames reveals that paint strokes differ markedly from one frame to another, raising questions about who executed what and whether some painted ornament may have been delegated to frame-makers or Whistler’s studio assistant Walter Greaves (1846–1930).

Whistler’s own rhetoric demands scrutiny. In 1878, during his libel trial against the art critic John Ruskin, he stated unequivocally that ‘The frame and the picture together are a work of art’.7 This implies that once the picture and its frame have been completed, they should have a unity that should not be altered or separated. In reference to his paintings Whistler once asserted that ‘the masterpiece should appear … as though it had been blown upon the canvas by a breath’, but the artist is known to have repeatedly scraped down his paintings and he also modified his frames.8 In one of his most informative letters discussing framing, he told his patron John Gerald Potter (1829–1908), the owner of Symphony in White No 2: The Little White Girl 1864 (YMSM 052; Tate N03418), that it should be reframed in time: ‘“The little White Girl” also ought to have a new frame. The old one is quite too weak for the picture – but there was no time to alter that – do order one’.9 It is not exactly known what ‘weak’ means here but Whistler believed the frame needed replacement.

The same letter goes on to say of the paintings ‘keep them under glass’ to protect them from the English climate and urban pollution – to achieve which might necessitate reframing. These instructions reveal not only Whistler’s lack of sentimentality towards the original surfaces of his frames, with regilding and repainting as a routine operation necessary for maintaining an intended appearance, but on occasions these changes were concurrent with him overseeing the cleaning and revarnishing of the accompanying painting. If, as was his ideal, the painting and frame were unified on completion as ‘a work of art’, this did not preclude changing the frame later, putting aside original intentions but creating a new unity from his own alternative frame designs and ornamentation.

These sources show that framing was a crucial part of Whistler’s art practice and that even at times of pressure – when he was subject to hostile criticism, dealing with the demands of exhibitions or seeking to counter environmental deterioration – framing was part of Whistler’s aesthetic and professional self-fashioning. The sections that follow examine how new technical evidence can be related to the meaning of ‘originality’ in a practice where renewal, repainting and regilding could serve as tools for sustaining the artist’s intention rather than negating it.

The frame around Nocturne: Blue and Gold – Old Battersea Bridge

The painting Nocturne: Blue and Gold – Old Battersea Bridge 1872–5 and its frame were presented to the National Gallery, London, by the National Art Collections Fund in July 1905, two years after the artist died. They were transferred later that year to the National Gallery of British Art, Millbank (renamed Tate Gallery in 1932). The gilded and painted frame around Nocturne: Blue and Gold – Old Battersea Bridge 1872–5, hereafter Battersea Bridge, was made by Foord & Dickinson, Carvers and Gilders, at 90 Wardour Street in London between 1876 and 1877 (fig.1). Their label was removed from it in 1983 and preserved. The painting was displayed at the Grosvenor Gallery in 1877, and no earlier frame is known. The complete work, painting and frame, was then fitted within a secondary glazed close-fitting hardwood overframe designed to protect the original frame (fig.2). At first sight the original frame appears less altered than it would have been by the fitting of glazing. It is possible that the assumption that the blue painted decoration was the artist’s own discouraged alterations to the frame by the gallery, such as fitting glazing doors or fit glass in the more conventional manner in the frame rebate. Certainly, the overframe‘s presence accounts for the reasonable state of preservation of the painted decoration.

The original frame is constructed from oak in three parts with a separate oak slip frame nearest the centre. The central painted flat frieze of the main frame sits between two bordering elements, into which it is rebated. The overall rounded form of these two is composed of a series of lines shaped into the timber, known individually as reeds and collectively as reeding. On the inner side of the central frieze the frame steps up by two reeds to a narrow channel which has segmented lines and dashes all painted in one direction on the vertical members but in opposition along the horizontal members. From this channel a succession of five reeds descends towards the slip frame, which is flat, about 12 millimetres wide, and is edged in a simple bevel above the painting. On the outer side of the central frieze the third element steps up by three reeds to a channel with painted blue lines and dashes, before the rounded form steps down through a succession of reeds to flat sides.

The frame was oil gilded – the process by which gold leaf is adhered to a frame using an oil-based adhesive – throughout. Typically, the wood is first sealed with a layer of animal glue, which stops the layer of oil size applied afterwards from sinking into the wood, with only two thin layers the oak grain remains visible (fig.3). The oil would be applied some hours beforehand, allowing its solvent to evaporate, until the optimum moment to lay the gold is chosen by the gilder. Added lead drier slows the drying time, allowing the brush marks to flatten out. If laid too soon the appearance of the gold is comparatively dull, too late and the gold adheres poorly. The painted ornamentation was applied to the gilding in blue oil paint. The pattern in the central frieze has been identified as seigaiha, a traditional Japanese pattern of overlapping concentric arcs (fig.4).10

The butterfly monogram with which Whistler often signed his work is depicted as negative space, in gold, with the blue paint as a surround and forming a circle on the right rail, just below centre (fig.5). This motif spreads into the two bordering channels, disrupting the segmented lines. Excepting the monogram, the painted ornament is symmetric around a central vertical axis, and the pattern runs harmoniously over the frame’s mitred corners, which was achieved by narrowing the forms as necessary. Throughout, the brushstrokes are applied with delicacy and finesse except for an unexplained smudged area on the lower rail (see fig.1).

In November 1878 Whistler sued Ruskin following the critic’s attack on the artist in his book Fors Clavigera (1877), where he accused Whistler of ‘flinging a pot of paint in the public’s face’.11 Ruskin was referencing Nocturne in Black and Gold: The Falling Rocket 1875 (YMSM 170; Detroit institute of Arts), but Battersea Bridge was also among paintings shown at the Grosvenor Gallery in 1877 and brought into the court at the request of Ruskin’s solicitors. Whistler believed the completed artwork, frame and painting, ‘would vindicate his position as an established accomplished artist’.12 He explicitly asserted authorship of the framing scheme: ‘The blue colouring on the gilt frame is part of the scheme of the picture. The blue spot on the right side of the frame is my monogram, which I place on the frame as well as the canvas; it balances the picture’.13 Ruskin’s accusation made Whistler’s control over ‘finish’ and total effect a public issue rather than a private preference.

In 1886 Whistler wrote to David Croal Thomson (1855–1930), art dealer and director of Goupil Gallery, New Bond Street, in the hope that he had purchased Battersea Bridge at an auction sale, commenting: ‘This you know is the very first example of the kind’.14 We cannot be sure if Whistler meant the painting or the frame or the work as a whole, but Whistler included it in his 1892 retrospective at Goupil.15 Yet examination of the frame around Battersea Bridge in advance of the 2026 Tate exhibition has brought into question the assumption that the visible blue pattern represents the original scheme. Traces of an earlier and worn seigaiha pattern are visible at the lower left corner of the flat frieze, and in some areas of reeding and in the frieze there are suggestions of an earlier gilding application (see fig.3). It is difficult to judge by eye alone the warmth or coolness of the gold in a partially revealed, earlier gilded surface because the oil size of the upper gilding scheme inevitably influences what is perceived beneath through gaps. There are, however, indications of a paler, more ‘lemony’ gold for the earlier scheme (see fig.3). Therefore, the largely visible scheme is a second layer of gilding and painting.

As described previously, the gilding appears as if applied directly to the oak, rather than to a chalk-based ground layer (whiting), leaving the wood grain easily discernible by eye, so the visibility of the wood grain must therefore have been an aesthetic choice. This aligns with a broader aesthetic discourse promoting ‘truth to materials’ in the decorative arts, articulated, for example, in Charles Locke Eastlake’s reformist writings on taste and finish.16 The traces of an earlier pattern raise the possibility that the visible pattern, which is undoubtedly painted almost everywhere on the frame with the same, extremely well-preserved paint, could in fact be a good second painting on a second oil gilding, albeit not necessarily with Whistler’s preferred blue tone or opacity of paint. The smudged area also defies explanation.

The frame today is seen as being in relatively good condition because the blue paint remains largely intact and the colour has not faded, although later animal-glue based toning layers have caused the gilding to lift in the interstices. The overframe was put on primarily to preserve what were assumed to be paint strokes of the artist, but its presence obscures the overall form of the frame and diminishes engagement with the frame as an object. The outer reeds at the front edges of the frame step down to the sides, fading into the dark shadow cast by the close overframe, which folds over the gilded frame. The overframe’s status as protective was rather diminished by treatment reports from 1983 that confirmed it had been fitted by screwing from the outside into the sides of the actual frame. Further alterations made in 1983 had included fitting additional screws on the top and lower sides for increased strengthening.

Out of the overframe, the actual frame sides are notably dark, the gold obscured. Newly gilded surfaces are painted with ‘gilder’s ormolu’ a thin transparent animal glue (protein) and shellac coating which provides protection, unifies the appearance disguising any overlaps of gold leaf, matts the surface and warms the overall tone. The glue content may have imbibed particulate dust from coal-heated and gas-lit London during the frame’s first thirty years of existence. Equally it could have resulted from further intentionally applied tinted glue-based layers used to obscure screw holes and wear at the corners, when the frame was displayed without the overframe. A cleaning test to the sides revealed, beneath this dark layer, a gold ‘lemony’ in appearance, different from the shade visible on the front between the paint strokes. The gold colour of the cleaned sides is like that on the slip frame, the innermost frame part closest to the glass or painting (fig.6).

For a frame conservator, the issue arises of how to treat such darkened coatings especially when an original ormolu layer is beneath – whether to view them as analogous to a picture varnish, which may be removed and replaced because it has discoloured, or as a coloured glaze to be retained. Since most such layers will include animal protein size, they cannot be separated one from another. Whistler himself would surely have wanted the surface cleaned (as shown in fig.4) and probably regilded.

Further alteration to the frame took place in 1983 when the previously mentioned slip frame was reduced to four narrower lengths of wood enabling a larger aperture, large enough to view the edges of canvas, so that the painting is no longer rebated into the frame in the conventional manner but appears to float. This is a twentieth century imposition on Whistler’s aesthetic.

We cannot know whether, had he had the opportunity, Whistler would have changed the frame entirely. From correspondence we know that he engaged in renewal of his frames and replacement. That the scheme we see today is not the first does not necessarily mean Whistler would not have approved; not honouring artist intent began with accession in 1905 with the imposition of the overframe. This may have been for the best intentions of securing the presumed paint strokes of the artist giving them primacy at the expense of the frame as an object. Continued sporadically recorded alterations have further distanced the frame from an original state.

The frame around Nocturne: Blue and Silver – Chelsea

The frame for Nocturne: Blue and Silver – Chelsea 1871, hereafter Chelsea, is described by MacDonald and Petri as original to the painting, designed by Whistler and possibly made by the decorator Frederick Fox (fig.7).17 It has a tripartite structure similar to Battersea Bridge. The central frieze painted with the blue seigaiha pattern and incorporates a painted ‘stinging butterfly’ (fig.8) on the right side, placed quite low on the oil gilding. This is bordered by channels decorated with what Day has described as a bamboo motif. MacDonald and Petri have suggested that Chelsea is one of the earliest frames with the seigaiha pattern.18 The appearance of the frame before recent treatment was murky and dull, with minimal tonal contrast between the paint strokes and the gilding. A heavy toning layer obscured the whole frame.

The paint on the central frieze is markedly different in appearance from that on Battersea Bridge. Here the paint strokes are thicker, but appear washed out, lacking both the delicacy of application and striking effect seen on the frame for Battersea Bridge. In addition, surface examination indicates regilding and repainting have taken place across much of the frame (possibly excluding the reeding), without the ‘scraping down’ that Whistler requested on occasions be undertaken by Grau.

Examination of cross-sections made from samples through the surface finish where it was damaged confirm two distinct decorative campaigns (fig.9). The first gilding layer was applied to the glue-sealed wood without whiting, so the wood grain would have been visible. The seigaiha pattern was painted in thick brushstrokes of pale blue paint made from lead white, cobalt blue, a trace of yellow ochre and a trace of pale yellow pigment that could be the pale cadmium yellow often found in Whistler’s paintings. This might have been mixed by Whistler. The second campaign involved sealing over with glue any dirt adhering to the first painting and gilding, then a coat of oil size containing yellow ochre being brushed on before the application of a pale, ‘lemony’ gold like that seen on the sides of Battersea Bridge. The second scheme was painted over the lemony gold, but what cannot be determined with confidence from cross-sections is whether the repainted pattern follows the original one in detail.

Close examination of the frame itself makes it clear that the second (and now visible) scheme of the seigaiha pattern has lost its colour almost completely where it is thin but retains a bluish-grey cast at the thicker beginning of each brushstroke. The paint medium has yellowed, and this yellowing prevails over the original blueness by the thin end of each brushstroke. The result is lines that are anaemic, almost transparent but with a brown cast despite their thickness. In fact, the appearance of thickness is in part due to underlying brushstrokes, which suggests that here the earlier pattern was largely followed. The anaemic pattern was made from a mixture of yellow ochre and Prussian blue in – likely – a natural-resin containing medium. This might have been a poor-quality commercial product, not one created by or for Whistler. To provide body and enable its thick application, it would have contained medium modifiers. Cross-sections confirm that its medium is significantly yellowed, and they also suggest that the later pattern lines were not applied with such thick, bodied paint as the original pattern. Why the first scheme had to be regilded is not explained in any known source. In a small number of locations where the gilding of the upper scheme has been lost, the blue of the first scheme can be discerned and it still looks blue, and much closer in tone to the blues in the painting. The cross-sections indicate that it would have had considerably more opacity, presence and blueness than the later painted pattern, because it contains a lot of opaque lead white (which the later paint does not).

The slip frame closest to the glass has a single scheme of gilding, consisting of whitish gold leaf on yellow oil size applied to thick chalk whiting bound in glue, making it distinct from both gilding schemes on the frame. An unusual feature of this slip is its asymmetry, wider in the vertical dimension while narrower top and bottom. This is probably explained in a letter written by Whistler to Walter Greaves during preparations for an exhibition at Dudley Gallery, London, in the winter of 1871: ‘Have they managed to fit in the little gold flat you know that Clay took down to the Gallery and that they wouldn’t let him put in the frame, but fixed it in themselves? Does it look all right? They have not taken off too much of the butterfly have they?’19 This letter provides a direct contemporary context for the frame’s surely unintentional asymmetric slip, and for the practical contingencies of fitting and adjustment in exhibition settings to an imminent deadline. The slip is most likely contemporaneous to the first gilding scheme. The second scheme appears more hurriedly applied: it could have been applied by one of the studio assistants, possibly under time pressure, for the 1892 Goupil Gallery exhibition.

Blue and Silver: Screen, with Old Battersea Bridge, the original screen

Blue and Silver: Screen, with Old Battersea Bridge is a hinged two-part folding screen possibly designed by Whistler as a Japanese-inspired piece of furniture, the structure of which was gilded, then ornamented with painted flowers (figs.10 and 11). On one side on darkened brown paper, his own nocturne-like rendering of the old Battersea Bridge with the Albert Bridge under construction spreads over the two parts while on the reverse are two paintings on silk of trees, plants in blossom, and birds. These have been recently attributed by Ayako Ono to the Japanese painter Ōzawa Nampo (born 1845).20 Whistler presumably purchased these paintings in London or Paris.

The screen was exhibited 1874 in the Flemish Gallery in Pall Mall, London, where it did not sell.21 He also hoped to sell it to his patron Frederick R. Leyland (1831–1892). Ultimately, he retained it. It can be seen in the background of his Paris studio when in 1893 he sat for a photographic studio portrait (fig.12). After 1901 when he closed his Paris studio, it returned to London. After his death it is seen again behind the surviving family of his deceased wife Beatrice Birnie Philip, all dressed in mourning clothes (fig.13). The screen was bequeathed by his executor and sister-in-law Rosalind Birnie Philip to the Hunterian, University of Glasgow, in 1958.

The screen’s construction is now complicated by a 1960s treatment that included the adhesion of both paintings to the opposite faces of two lengths of board, and the fitting of glazing over the silk painting by Ōzawa Nampo. The current hinging of the original screen appears not to be contemporaneous to the painting, and several lower rails may have been lost, since the empty mortices remain in the uprights. The original hinging is unknown but potentially it is only the current leather straps threaded through the frame that are non-original rather than the methodology. Additionally, the gilded horizontal top member directly above the bridge and moon is also likely a replacement post-dating the artist. The 1960s intervention has put stress on the wooden frame, which now supports much greater weight than it was designed to, and the screen is too fragile to travel. For the remaking of the screen for the 2026 exhibition, the construction of the framework was undertaken following an examination of the original framework and a re-evaluation of the nocturne’s 2003 technical study.22

The gold leaf used by Whistler for the pine framework is a gold and silver alloy, laid using an oil mordant, now dirty and scuffed and appearing greenish in tone. It was applied to a thin, glue-based preparatory layer on the softwood frame, made from chalk and extenders with traces of Prussian blue, yellow ochre and sienna, sealing the wood to stop the oil mordant from sinking. Viewed from the Whistler side of the screen the butterfly monogram is on the right in a circle, placed level with the bridge and orange moon but with the insect body central on the outermost edge and the wings stretching round towards both paintings. The glue-based preparatory layer is thick and impasted beneath Whistler’s butterfly monogram, so he must have determined its position and the form of the butterfly, likely applying this in person before the gilder began work.

The flowers were not prepared or marked out in anyway, and therefore may have been freely painted, although not necessarily by Whistler. They look like abstracted nasturtiums, the majority with five free floating petals and with each petal having three lobes. For visual variety there are flowers of different sizes; some appear to be floating away off the edge of the framework; there are single petals and two or three petalled flowers; in places there are groups of single, coloured flowers; there are no flower stalks but apparently randomly made horizontal dashes throughout. The paint for the nocturne is very matt, and it is likely to be protein-based – perhaps also animal glue – but the painted flowers on the frame are painted in oil, from visual examination. While undertaking the creation of the replica it became apparent that only one hand could have been involved in painting the framework.

Both blues for the original flowers were made with lead white, Prussian blue mixed with yellow ochre, small amounts of viridian, sienna and a fine-grained black pigment, along with extenders, the paler shade having more lead white. These were mixed together roughly, leaving large lumps of blue and yellow in the paint. It is possible that Whistler used decorators’ pigments to mix this paint: a letter from the builder who refurbished his home in Lindsay Row in the 1870s, listing decorating materials and costs, mentions all the colours consistent with the pigments identified in the screen’s painted ornament. The paint mixture is quite a complex and specific one, however, and perhaps over-complex for a decorators’ paint, which would have been formulated with material costs firmly in mind.

Remaking of the framework for Blue and Silver: Screen

Only the framework of the screen was recreated, using materials comparable to the original. It is not a replica of the original as it appears now, but a materials-based reconstruction of how it may have looked when newly made (fig.14). Whistler’s nocturne and Ōzawa Nampo’s paintings were substituted with digital prints. Tulip wood was used for the frame, the framework constructed with mortices and tenons and a further single lower rail beneath the panels connecting the uprights but not replicating the empty mortices on the original. The leather hinges were copied, but improved to facilitate packing and shipping of the replica.23 A cross-section through the original would show, that the width of timber around the Whistler is 3 to 5 millimetre wider than on the equivalent Nampo side which has a narrow beading 3 to 5 millimetre wide, but the decision was taken to shape the timber symmetrically on a spindle moulder.

A thin preparatory layer made of chalk, clay and brown umber in animal glue medium was then applied to mimic the warm brown colour of the aged pine wood visible on the real frame through gilding losses. After studying the original, the raised monogram was replicated here using acrylic paint and medium thickened with chalk and painted to mimic as closely as possible the original paint strokes using life-size prints of the screen (from the same digital images used for printing the two sides). The surface was then oil-gilded with 16-carat green gold identified on the screen itself through observation, notably where the artist’s paint had fallen away leaving paler patches than between the flower petals. This echoed the tonality of the original gilding on Battersea Bridge and Chelsea. The gold was protectively coated with gilder’s ormolu made with parchment glue and unbleached shellac, and then a yellow tinted varnish (fig.15). It is still more ‘lemony’ than ‘buttery’ in tone; flowers were then painted free-hand using acrylic with gloss medium in two shades of blue, to mimic the original colours and paint texture (fig.16).

Whistler’s reframing in the 1890s

By the 1890s Whistler had become increasingly concerned about the condition of his paintings, such as Nocturne Blue and Silver: Cremorne Lights, and the alteration of their colours, glazing became a necessity. He wrote in March 1892 to Potter: ‘I take this occasion to beseech you always to keep them under glass – The climate and smoke of England make it absolutely imperative’.24 This insistence sits within a wider late Victorian discourse on industrial pollution, exemplified by Ruskin’s lectures The Storm-Cloud of the Nineteenth Century (1884), which articulated the danger to art from urban pollution.25 Read together, Ruskin’s environmental alarm and Whistler’s practical instruction to glaze paintings converge on the shared reality that London’s heavily polluted air seriously altered and damaged painted and gilded surfaces in the late nineteenth century.

The Grau style frames on four Tate paintings

The four frames at Tate categorised under the name Grau are those around Symphony in White, No. 2: The Little White Girl, Crepuscule in Flesh Colour and Green: Valparaiso 1866 (YMSM 073; Tate N05065; fig.17), Nocturne: Blue and Silver – Cremorne Lights 1872 (YMSM 115; Tate N03420) and Nocturne: Black and Gold – The Fire Wheel 1875 (YMSM 169; Tate N03419; fig.18). All four frames were treated before the exhibition in 2026. However, only two of these frames were made by Grau himself, around the second and fourth paintings, hereafter Valparaiso and The Fire Wheel. Those around the first and third painting, hereafter Symphony in White No.2 and Cremorne Lights, appear to be by two different unknown makers.

Of Grau, Whistler wrote to a dealer in June 1892: ‘He is the only one who has the true pattern of my frame – Tell him that the gold must be the pale yellow soft gold like the gilding of my Mother’s frame’.26 The Fire Wheel was later regilded, and this non-original scheme was recently removed to reveal Grau’s gilding, which would be described today as soft and warm in tone and distinct from the cooler tones and ‘lemony’ colouration of the earlier frames. The colour of the gold for these frames was different from those used on the painted frames of Battersea Bridge and Chelsea, which were cooler or lemony. Although Whistler described the colour precisely, exactly what he envisaged by this can only be estimated today by examining the frames, now more than 130 years old.

Valparaiso may have been the prototype of a Grau frame (fig.19), and in form it is the least altered of the Tate quartet. It is inscribed in pencil by Grau on its reverse ‘FH Grau, London, 1891’, most likely the first frame Grau made for Whistler. Soon after the exhibition at Goupil the painting was sold at auction to the collector W. Graham Robertson (1867–1948) but a letter from Whistler to Robertson written shortly thereafter suggests Valparaiso was not in the Grau frame yet. Whistler suggested Robertson ‘send for Grau – and tell him to make you at once one of my beautiful new frames … Of exactly the same gold as that he has used for me lately … Your picture in the new frame will look five times as stately and beautiful’.27

The curvature of the overall profile shape of the frames made by Grau himself is distinctive.28 The gilded slip frames for The Fire Wheel and Valparaiso are more subtly made than a casual glance would suggest. But, crucially, from the front edge to each successive channel, there is a step up in height, so that neither of the two gilded reeds would be rubbed by the glazing above, which is separated from them by a 2.5 or 5-millimetre gap.

The frame around The Fire Wheel was made in 1892, for the Goupil exhibition, although it was not ‘signed’ by its maker. The Grau frames for The Fire Wheel and Valparaiso have different histories but are similar in design concept, being constructed with a sloping profile to the reeding such that their original glazing does not contact the tops of the reeds and wear off the gilding, at least where they are visible to the gallery viewer. Both frames incorporate slips.

The slip frames that surround the paintings, as viewed from the front, vary across the Tate quartet and show differing success in balancing elegance and function, assuming the paintings were glazed from the time of their making, which today is unknowable since this was not recorded before they were modified. For example, the slip for Symphony in White, No. 2, is straight across, that is, its reeding all lies in the same plane, so by its design the high points of the gilding would rub against the glass. For Cremorne Lights, the slip avoids any damage to the gilding against the glazing despite the reeds and channels being in one plane, as there is a final step up over an inelegant rounded edge into the rebate. These are details unlikely to be noticed by a viewer today, but they do illustrate the degree to which Whistler designed the frames not only to accommodate glazing but with the preservation of the frames’ finish in mind, and also the care which Grau took to make such frames. This solicitude was not necessarily followed by other makers.

Subsequently, three of the frames – those around Symphony in White, No. 2, The Fire Wheel and Cremorne Lights – were altered to have glazing doors, quite drastically to achieve this result. A glazing door allows for the glazing to be fitted and removed from the front of the frame. There are different possible arrangements. They allow the glazing to be taken out and cleaned inside, or taken out to show the painting without glazing or taken out to lessen the weight when the frame is taken down, and reduce the risk of the glaze breaking and damaging the painting. The effect is visually as well as physically to push the painting back in the frame, making the distance between viewer and painting feel greater, although it is only a moderate measurable distance. Valparaiso was not altered in this manner and has been the structural model to which the other frames have been restored for the 2026 exhibition. Valparaiso had been regilded at an unknown date, but this regilding had flaked badly, and the decision was made to remove this upper layer by dry stripping, to reveal a gilding scheme that is consistent with the ‘yellow soft gold’ preferred by Whistler.

In art historical terms, the Grau frames also align with contemporary reformist ideas about subdued gilding and refined finish. Eastlake’s Hints on Household Taste in Furniture, Upholstery and Other Details (1868) advocated restraint in ornament and careful handling of surface and tone.29 Whistler’s preference for ‘pale yellow soft gold’ reads as a highly personal, artist-driven iteration of that reformist palette.

Conclusion

Because gold can be alloyed with other metals, such as silver and copper, to create different shades of leaf, it can be difficult to envisage what a given individual meant when describing the ‘colour’ of gold. It is equally challenging to assess what was perceived originally, as compared to what we see on the surface today after more than 150 years of wear, handling and dirt accumulation. In this article we have used the word ‘lemony’ to describe the gold ‘colour’ on the two early Tate Nocturne frames. For our ‘lemony’, might Whistler have perceived and stated ‘green’ – In a letter that references Nocturne: Blue and Silver – Bognor 1874–6 (YMSM 100; National Museum of Asian Art, Washington), he described how the painting should ‘be framed in pale green gold with blue pattern’.30 It becomes difficult to honour the harmonies sought by Whistler when individuals may interpret gold ‘colours’ supplied today, such as citron, lemon or green, in different ways. If original colour harmonies in the frames are lost through wear, loss or the fading of paint, both removal and application of toning layers could distort original intention.

The Hunterian Screen was made in the same period as the two Tate nocturnes, and it is plausible that the gold colours on all would have been similar. Close examination of all three supported this hypothesis. The remaking of the screen provides a unique aid to visualising Whistler’s aesthetic preferences for a gilded frame in the early 1870s.

The frames for Symphony in White, No. 2, Cremorne Lights, Valparaiso and The Fire Wheel would have been near identical in appearance when first made, as regards the colour of the gold and the quality of their making. They represent Whistler’s final design concept but with differing detail, the lens through which he wished posterity to view his paintings, with a gold colour softer and warmer than the tones he previously favoured. Significantly, they allowed for glazing to protect the paintings from London pollution. Those made by Grau were constructed carefully to prevent the glazing from rubbing gold off the highpoints of the gilded reeds, the better to retain their original appearance.

The 2026 exhibition has provided a useful nexus for re-evaluating Whistler’s frames for the Tate paintings. The revealed gilding schemes are consistent with Whistler’s known views on framing and shed a welcome light on the actual appearance these frames would have had when new. They provide understanding that could enable future treatment of the painted ornament, and remediation of the non-original overframe on Battersea Bridge.