Joan Miró

A sourced technical report on his oils on canvas, 1920s to 1970s. Every claim cites a source key; the sources are listed at the end. "Uncertain" marks claims that are unverified, disputed or taken from a single object; each piece of evidence is dated. The core evidence is a survey of the tubes and paints left in his Mallorca studios, with analyses of 1970s paintings [GL23; HS23; POL24], a conservation report on three canvases of 1962 [AIC19], three Tate catalogue entries [TATE27; TATE38; TATE49] and his recorded statements, all read in full. A study of a canvas of 1925–27 was read only through press summaries [FJM25]; analyses of canvases of 1938 and 1949 only through a pigment table in the studio survey and search excerpts [GL23 Table 1; OD06; BURN91]; the museum's account of a canvas of 1925 through search excerpts and an encyclopedia summary [MOMA25].

Summary

He painted mostly on canvas, some reused, some left raw. A large canvas of 1925 was sized with animal glue and painted with thin, dripping paint; linen of 1962 was heavily sized and given a glue-bound ground [MOMA25; AIC19]. He said he never began on a canvas as it came from the shop, but with accidents, such as wiping his brushes on it [NEVER]. A canvas of 1927 has a water-based house-paint ground under motifs in oil [TATE27]. From the 1940s he often prepared canvases with a casein product [GL23 p.26]; from 1960 he added vinyl and acrylic paints, often for grounds, without giving up oil [GL23 p.22; POL24 "Introduction"]. Drawings came first, sometimes closely transferred [THY; POMP]. He kept many canvases going at once, some for years [GARD]. His oil colors were mostly cadmiums, cobalts and earths, with cerulean, viridian and ivory black, and lead, zinc and titanium whites [GL23 Table 1; SAL03]. The documented pigments are listed in §4.

Terms. Ground/priming: the preparatory layers between canvas and paint. Size: animal glue brushed on the bare canvas to seal it. Gesso: a white ground of chalk in glue. Casein: a milk-protein glue used as a binder. Tempera: paint in egg or another water-based binder. Vinyl paint: a water-based paint bound in polyvinyl acetate (PVAc). Enamel: a house paint in an alkyd resin. Metal soaps: compounds that form when metal ions from a pigment react with the oil. Impasto: thick, raised paint.

1. Supports

• Canvas, at every size. Canvases of the 1920s run from 55 × 46 cm to 250.8 × 200 cm; one of 1949 is 81.5 × 66 cm [THY; MOMA25; TATE49]. From 1956 a large Mallorca studio let him work "on large-format artworks, and on many works simultaneously" [HS23 "Introduction"]. Canvases of 1961 and 1973–74 measure about 270 × 355 cm, three of 1962 about 366 × 347 cm [POMP; HS23 "Materials and methods"; AIC19 p.63].
• Linen. The 1925 canvas is a very fine linen [MOMA25] (read through search excerpts); those of 1962 are "open-weave linen" [AIC19 p.63].
• Reused canvases. A canvas of 1925–27 is painted over another painter's portrait of his mother. He reused canvases bearing academic portraits in the 1940s–50s and landscapes in the 1960s–70s [FJM25]. Read through press summaries.
• Raw canvas. For one canvas of 1966 he used unprepared canvas, "applying charcoal and white paint directly on the cloth" [GL23 p.26].
• Cut and burned, 1973. Five canvases were cut with a knife and punctured, then painted, soaked in petrol and set alight [TETC].
• Other supports (outside this note): copper (1935), hardboard (1936), plywood (1977), sandpaper, tarpaper and burlap [SAL03; MAS; POL24 Fig. 2; SCHIRN]. The Constellations of 1940–41 are "gouache and oil washes on paper" [MOMA19].

2. Grounds

• Never a shop canvas. "Never, never do I set to work on a canvas in the state it comes in from the shop. I provoke accidents … Then I prepare a ground by, for example, wiping my brushes on the canvas. Letting fall some drops of turpentine on it would do just as well" [NEVER]. Uncertain: widely quoted; the interview was not identified.
• Glue size, 1925. The museum calls the 1925 canvas "unevenly primed"; its conservator reports glue sizing, then very dilute paint from the top [MOMA25]. Read through search excerpts and a summary.
• Water-based blue, 1927. A canvas of 1927 has a "water-soluble background and motifs in oil." The background "appears to have been painted with the kind of water-soluble blue paint that is frequently used on the outsides of houses in Spain and Portugal," as were others that year [TATE27]. The museum now gives tempera and oil [TATE27].
• Zinc white over an old painting, 1925–27. Over the portrait he laid a preparation of zinc white and barite (barium sulfate), then the blue. Strokes run vertically in the preparatory layers, horizontally in the blue [FJM25]. Read through press summaries.
• Lead white on copper, 1935. Paintings on copper have lead white in a drying oil under cadmium red and yellow in oil [SAL03]. Read through a search excerpt.
• Casein, 1940s on. "Until the introduction of synthetic primers, Miró frequently used Case-Arti to prepare his canvases"; his working notes of 1941–42 name it [GL23 p.26, translated]. A packet in the studio is calcium carbonate, perhaps with kaolin, in casein [GL23 pp.25, 27].
• Refined grounds, 1949. A series of 1949–50 is "noteworthy for the diversity and extraordinary refinement of their grounds" [TATE49, quoting his cataloguer].
• A brushed blue ground, 1961. He painted "first the ground, all blue," not "like a house painter: all the movements of the brush, those of the wrist, the breathing of a hand came in too" [POMP, quoting a 1961 interview; translated].
• Glue and gesso, 1962. The canvases "were originally sized using a large amount of animal glue," which sealed the weave, under a glue-bound gesso ground [AIC19 p.63].
• Vinyl and oil grounds, 1970s. He often used vinyl paints for grounds [GL23 p.22]. Two canvases of 1973 have black grounds in a vinyl paint from his studio [GL23 p.23]; one of 1974 has "a diluted oil with ivory black pigment" [GL23 p.26, translated].

3. Drawing and lay-in

• Accidents first. A picture could start from "a tiny thread sticking up from the canvas, a falling drop of water, or a print made by my finger on the shining surface of a table" [GARD].
• Drawings transferred, 1927. "First he sketched the image automatically in a drawing and later transferred it carefully to canvas"; one canvas is "an almost literal rendering" of a drawing [THY]. Read through a summary.
• Signs over a loose ground, 1925. Over the poured ground he added "schematic lines and seemingly familiar shapes … planned in preparatory studies" [MOMA25].
• Tiny sketches, 1960. The large blue canvases began as tiny sketches on scraps of paper [POMP].
• Charcoal and chalk. Canvases of 1966 and 1973 are catalogued with charcoal, one of 1974 with chalk [GL23 pp.24, 30].
• Color named on the drawing. Preparatory drawings in Mallorca carry the note "bleu céruléum" [GL23 p.21].

4. Pigments

The survey counted 175 oil tubes in his two Mallorca studios: 36 in the main studio, most used, and a box of 139, apparently an order of July 1966, mostly unopened [GL23 pp.15–16]. Its table adds the colors he names in letters and notes and pigments found by analysis [GL23 Table 1]. Many tubes of a color may mean a favorite, or a color left unused [GL23 p.14].

Pigment	Evidence	Notes
Lead white	Ground on copper, 1935 [SAL03]; four tubes of "silver white" [GL23 p.21]	"No other references" to it [GL23 p.21]
Zinc white	His preparation of 1925–27 [FJM25]; a canvas of 1949; a 1920 letter [GL23 Table 1]
Titanium white	Canvases of 1938 and 1949 [GL23 Table 1]; vinyl paint of 1977 [POL24 Table 1]
Lithopone	Canvas of 1949 [GL23 Table 1]	Single object
Cadmium yellow, lemon grade	1918 letter (below) [HS23 "Introduction"]; 1970s canvases and palettes, matching five tubes of "Cadmium Yellow Lemon No. 1" [HS23 "Discussion"]	Zinc-bearing, poorly crystalline [HS23 "Results"]
Cadmium yellow	Paintings of 1935 (printed 1933) and 1938 [GL23 Table 1]; a little in the blue of 1925–27 [FJM25]; Spanish-brand tubes [HS23 "Results"]	All his yellow and orange oil tubes are cadmium [GL23 p.20]
Cadmium orange	Canvas of 1938; three tubes "cadmium yellow orange No. 5" [GL23 Table 1]
Cadmium red	Canvas of 1938; letters of 1918 and 1920 [GL23 Table 1]; fourteen tubes [GL23 p.20]
Vermilion	Canvas of 1938 [GL23 Table 1]	Single object
Red lake	Canvas of 1938; one tube of "crimson lake" [GL23 Table 1]	Type not stated
Iron oxide orange	Canvas of 1949 [GL23 Table 1]	Single object
Red earths	"Pozzuoli earth" and red ochre in letters of 1918 and 1920; one tube of each [GL23 Table 1]
Raw sienna	1920 letter; one tube [GL23 Table 1]
Burnt sienna	Working notes of 1941–42 [GL23 Table 1]
Raw umber	Two Spanish-brand tubes, one label uncertain [GL23 Table 1]	No use documented
Cobalt blue	The blue of 1925–27, "cobalt blue with a little cadmium yellow" [FJM25]; canvas of 1949; 1920 letter; six tubes [GL23 Table 1]
Cerulean blue	43 tubes, the most of any color, 41 of them in the 1966 box; named on drawings [GL23 p.21, Table 1]	Mallorca period
Prussian blue	Notes of 1941–42 and a Mallorca manuscript; four tubes from three makers [GL23 Table 1]
Synthetic ultramarine	Two tubes [GL23 Table 1]	No use documented; "synthetic" is an inference
Viridian	Canvases of 1938 and 1949; "emerald green" in a 1920 letter and the 1941–42 notes; tubes labelled "emerald green (viridian)" [GL23 Table 1]	So the French "vert émeraude" meant viridian (inference)
Phthalocyanine green	"Monastral green," canvas of 1949 [GL23 Table 1]	Single object
Cobalt yellow, zinc yellow	Canvas of 1949 [GL23 Table 1]	Single object
Cobalt violet	Canvas of 1938; two tubes [GL23 Table 1]
Ivory black	Canvases of 1949, 1974 and 1978; letters of 1920, 1924 and 1934; notes of 1941–42; nine tubes [GL23 Table 1, pp.21, 26]	The only black oil in the studios [GL23 p.21]
Lamp black	Vinyl black of the 1973 grounds; an enamel can of 1965 [GL23 pp.22–23, 25]	Not in oil

• His own advice, 1918. "Instead of chrome, I recommend … cadmium lemon, it is splendid… Chrome, next to cadmium lemon, is a dull and lightless colour" [HS23 "Introduction"]. That year a fellow painter told him of a new cadmium red from an older Paris firm: "When I go to Barcelona, I'll look for it" [GL23 p.20, translated].
• Cadmium green. The studios hold 26 tubes of "cadmium green," which "is not a pigment in itself, but a mixture of cadmium yellow and viridian" [GL23 p.16]. A photograph of 1969 shows him using one [GL23 p.20].
• Names without a pigment. His letters and notes also name strontium yellow and a "yellow orange" (1920), a green sold under a Venetian painter's name (1918, 1941–42), a green and a brown named after a Barcelona paint maker (1941–42) and a "bitumen black" [GL23 Table 1].
• Chromium in yellows. A 2025 scan of Mallorca works found yellows containing chromium [MAL25]. Uncertain: no pigment was named.
• Makers. His letters of 1918–34 name an older Paris firm. Most Mallorca tubes come from a small Paris specialist maker, in whose house he stored his works from 1939 to 1956; a few from the older firm, a London colorman, a Dutch maker and a Spanish brand [GL23 p.16, Table 1].

5. Media and varnish

• Oil, with excursions. Oil was his main medium until the 1960s, with egg tempera in the 1920s and occasional enamel from the 1930s; acrylics came in the 1960s, "but without displacing oil" [GL23 p.15, translated].
• Enamel, 1930s. Paintings of 1935 are catalogued as oil and enamel; notes of 1937 and 1940 name matt enamels [GL23 pp.21–22]. Hardboard paintings of 1936 combine "oil and enamel paints, casein, tar, sand, and pebbles" [MAS]. Read through a summary.
• Vinyl and acrylic, 1960s–70s. His first note naming a Spanish vinyl paint is of 1960 [POL24 "Introduction"]. In 1974 he said he used acrylics "because of their rapid drying properties," but valued the richness of "slow-drying" oil; he often combined them [POL24 "Introduction"; GL23 p.22]. They let him try "drips and action painting" [GL23 p.22, translated].
• Oils. Oil paint on canvases of 1973–78 is cold-pressed linseed in one, stand oil (heat-thickened linseed) in another and probably safflower in a third. The specialist maker's tube is in safflower or sunflower oil with a pine resin; the Spanish tubes are in linseed oil [HS23 "Results"].
• Resin on the palette. Paint on his palettes holds linseed oil and a pine resin, perhaps added by him to thicken it [HS23 "Results"]. Uncertain.
• Thinned oil. The 1974 black ground is diluted oil [GL23 p.26]; the 1925 canvas was painted with very dilute paint [MOMA25].

6. Layers and application

• Poured, flung and soaked, 1925. He applied the paint by "pouring, brushing, and flinging," "so that the paint soaked into the canvas in some places while resting on the surface in others" [MOMA25]. Drips run down from the top, "a very controlled accident" [MOMA25]. Read through search excerpts and a summary.
• Drips on blue grounds, mid-1920s. He "dripped and splashed paint onto blue grounding in order to create the impression of old, weather-worn walls." Brown grounds form "the second largest group" after blue [SCHIRN].
• Two layers over an old portrait, 1925–27. Seven layers: a lead-white ground and four paint layers (taken here to be the portrait's; inference), then his white preparation and the blue. He left the relief of the portrait's brooch and earrings showing [FJM25]. Read through press summaries.
• Accents on black, 1938. On a painting of 1938, "accents of colour (red, yellow and white)… have been painted on top of the black ground" [TATE38].
• Signs on blue, 1961. On one large blue canvas the signs are only black and red points, lines and slashes [POMP].
• Impasto and thin paint. A sample of 1978 came from "the impasto under the star" [GL23 p.30, translated]; in other 1970s works the yellow is thin [HS23 "Materials and methods"].
• Fire, feet and fingers, 1973. Paint and petrol were poured on and lit, twice, with a wet mop and a blowtorch to control the fire; he walked on the canvas and painted with his fingers [TETC].

7. Working sequence and reworking

• Many canvases, for years. "I'm always working at a great many things at the same time." "I work for a long time, sometimes years, on the same canvas. But all this time, there are periods, sometimes quite long, when I'm not doing anything with it" [GARD].
• Slow series, 1949. A series of 1949 "was executed very slowly" [TATE49, his letter of 1954].
• Meditation first, 1961. "I took a long time to make them. Not to paint them, but to meditate them" [POMP, translated]. "'Perfecting' the ground put me in a state to continue the rest" [POMP].
• Planned phases, 1973. His notes for the burned canvases list "I: Pour colours, II: Cuts, III: Tear and hang it, IV: Black Spots" [TETC].
• Painting over. On a canvas of 1962 he appears to have repainted much of the background; a halo of earlier work shows [AIC19 p.65].
• Floor and wall. A photograph of 1969 shows works in progress on the floor [GL23 p.20]. The 1925 drips run from the top, so that canvas stood upright at least for a time (inference) [MOMA25].

8. Condition and aging

• Fading cadmium yellow, 1970s. In 26 works the yellow has faded, turned powdery or grown a whitish crust as the poorly crystalline cadmium sulfide turns to colorless sulfates [HS23 "Introduction", "Results"]. Strong light and winter humidity above 80% in the studio were "a major factor"; a palette kept in a case stayed bright [HS23 "Discussion", "Conclusions"].
• Zinc soaps. A canvas of 1938 shows flaking and tiny spherical protrusions, with zinc soaps [OD06] (read through search excerpts). Zinc stearate occurs in the 1974 black oil ground [GL23 p.26].
• Black spots on paintings on copper of 1935 were studied separately [SAL03]. Not read.
• Glue-sized canvases. The canvases of 1962 slacken above 55% relative humidity and contract hard below 25%. Within ten years they flaked and cracked, and a technician, with his permission, glued their backs [AIC19 pp.63–64].

9. The tubes here

The historical names in this note are mapped to these tubes: lead (silver) white → lead white; zinc white → zinc white; titanium white → titanium white (a new tube); cadmium lemon → pale cadmium; cadmium yellow → cadmium yellow; cadmium orange → deep cadmium; cadmium red → cadmium red; vermilion → vermilion; red and "crimson" lake → rose madder (a proxy: the lake was not identified); iron oxide orange → Mars orange; Pozzuoli earth and red ochre → red earth; raw sienna → raw sienna; burnt sienna → burnt sienna; raw umber → raw umber; cobalt blue → cobalt blue; cerulean → cerulean blue; Prussian blue → Prussian blue; ultramarine → ultramarine blue; viridian (sold as "vert émeraude") → viridian; cobalt violet → cobalt violet; ivory black → bone black (both charred animal matter: ivory was "occasionally adulterated with bone black" [SAL §293]). Cadmium green is mixed from a cadmium yellow and viridian. Vermilion, the red lake, iron oxide orange and raw umber rest on one painting or on tubes alone, titanium white on two analyses read second-hand (§4). Lithopone, phthalocyanine green, cobalt yellow, zinc chromate yellow and lamp black have no tube.

10. At this easel

• Support. Plain-weave linen with ground layers laid from the tubes.
• Paint. Pigment from the tubes with an oil-medium share.
• Taking paint off. A brush picks up wet paint as it passes; that is how paint comes off the canvas.
• Everything else the easel does is in the easel guide.

Sources

• [GL23] Gómez Lobón, M. et al., "Investigación de los pigmentos utilizados por Joan Miró durante su etapa artística en Mallorca," in Conservación de arte contemporáneo: 24ª jornada (Madrid: Museo Nacional Centro de Arte Reina Sofía, 2023), pp.13–36. Table 1 (pp.17–19) lists his oil tubes by color and maker, colors named in his letters, notes and manuscripts, and pigments found in earlier analyses of paintings of 1935, 1938 and 1949. Full text read; quotations translated from Spanish. https://recursos.museoreinasofia.es/conservacion_de_arte_contemporaneo_24a_jornada_red.pdf
• [HS23] Gomez Lobon, M. et al., "A study of cadmium yellow paints from Joan Miró's paintings and studio materials preserved at the Fundació Miró Mallorca," Heritage Science 11 (2023), art. 145. Cited by section. Full text read. https://doi.org/10.1186/s40494-023-00987-4
• [POL24] Gomez Lobon, M. et al., "Unveiling the Composition of La Pajarita PVAc-Based Paints in Joan Miró's Studio and in Three Artworks from the 1970s," Polymers 16:22 (2024), 3146. Cited by section. Full text read. https://doi.org/10.3390/polym16223146
• [AIC19] Iaccarino Idelson, A. et al., "The Conservation History and Treatment of Three Large-Scale Paintings by Joan Miró," AIC Paintings Specialty Group Postprints 32 (2019), pp.63–74, on three canvases of 1962. Full text read. https://zenodo.org/records/18300873
• [FJM25] Fundació Joan Miró, Barcelona, technical study of a canvas of 1925–27 painted over an earlier portrait, announced March 2025. Read through press summaries: Ara, 8 April 2025, https://es.ara.cat/cultura/arte/retrato-madre-joan-miro-sacrifico-librarse-tradicion_1_5329257.html ; Diari de Tarragona, https://www.diaridetarragona.com/cultura/225567/los-investigadores-de-la-fundacio-joan-miro-han-descubierto-un-retrato-de-la-madre-del-artista-bajo-la-obra-pintura-ga23451610.html . The foundation's release was behind a bot check: https://www.fmirobcn.org/en/foundation/press/331/researchers-at-the-fundacio-joan-miro-discover-a-portrait-of-joan-miro-s-mother-beneath-the-work-painting-c-1925
• [OD06] O'Donoghue, E. et al., "Dictated by media: conservation and technical analysis of a 1938 Joan Miró canvas painting," Studies in Conservation 51, suppl. 2 (2006), pp.62–68. Not read; abstract through search excerpts, pigment list through [GL23] Table 1. https://doi.org/10.1179/sic.2006.51.Supplement-2.62
• [SAL03] Salvadó, N. et al., "Nature and origin of black spots found on Miró paintings: a non-invasive study," Analytica Chimica Acta 479 (2003), pp.255–263. Not read; known through a search excerpt and [GL23] Table 1. https://doi.org/10.1016/S0003-2670(02)01530-1
• [BURN91] Burnstock, A., unpublished PhD thesis, Courtauld Institute of Art (1991), pp.265–268, analysis of the Tate canvas of 1949. Known only through [GL23] Table 1.
• [TATE27] Alley, R., Catalogue of the Tate Gallery's Collection of Modern Art other than Works by British Artists (London 1981), pp.524–525, entry T01318 (1927). Read as archived in 2012: http://web.archive.org/web/20120720194851/http://www.tate.org.uk/art/artworks/miro-painting-t01318/text-catalogue-entry ; current record: https://www.tate.org.uk/art/artworks/miro-painting-t01318
• [TATE49] Alley 1981, pp.525–526, entry N06007 (1949), quoting his letter of 17 June 1954. http://web.archive.org/web/2015/http://www.tate.org.uk:80/art/artworks/miro-women-and-bird-in-the-moonlight-n06007/text-catalogue-entry
• [TATE38] The Tate Gallery 1984–86: Illustrated Catalogue of Acquisitions (London 1988), pp.533–536, entry T03690 (1938). http://web.archive.org/web/2015/http://www.tate.org.uk:80/art/artworks/miro-a-star-caresses-the-breast-of-a-negress-painting-poem-t03690/text-catalogue-entry
• [TETC] Jeffett, W., "Burn, canvas, burn," Tate Etc. 21 (Spring 2011), on the burned canvases of 1973. https://www.tate.org.uk/tate-etc/issue-21-spring-2011/burn-canvas-burn
• [POMP] Centre Pompidou, collection record for Bleu II (1961), quoting his interview in L'Œil 79–80 (July–August 1961), p.18. Quotations translated from French. https://www.centrepompidou.fr/fr/ressources/oeuvre/crdy9X
• [MOMA25] The Museum of Modern Art, on The Birth of the World (1925): its description as quoted at https://en.wikipedia.org/wiki/The_Birth_of_the_World , and its paintings conservator's audio commentary, read through search excerpts, https://www.moma.org/audio/playlist/296/3853
• [MOMA19] The Museum of Modern Art, press release for Joan Miró: Birth of the World, 20 February 2019. https://assets.moma.org/documents/moma_press-release_479263.pdf
• [THY] Museo Nacional Thyssen-Bornemisza, collection text on Painting on White Ground (1927). Read through a summary. https://www.museothyssen.org/en/collection/artists/miro-joan/painting-white-ground
• [SCHIRN] Schirn Kunsthalle Frankfurt, "Joan Miró. Painting walls, painting worlds," Schirn Mag, 8 February 2016. https://www.schirn.de/en/schirnmag/joan-miro-painting-walls-painting-worlds-context-en/
• [MAS] Encyclopedia entry "Paintings on Masonite," citing R. M. Malet, Joan Miró (Barcelona 1983), and the Museum of Modern Art's 2008 press release. Read through a summary. https://en.wikipedia.org/wiki/Paintings_on_Masonite
• [GARD] His interview with a French critic, "Je travaille comme un jardinier," XXe Siècle, 15 February 1959; English edition Joan Miró: I Work Like a Gardener (Princeton Architectural Press, 2017). Read through excerpts: https://designobserver.com/feature/joan-miro-i-work-like-a-gardener/39682 ; https://www.themarginalian.org/2015/09/17/i-work-like-a-gardener-joan-miro/ ; https://www.anothermag.com/art-photography/10378/i-work-like-a-gardener-joan-mirs-most-compelling-quotes
• [NEVER] His statement beginning "Never, never do I set to work on a canvas in the state it comes in from the shop," widely quoted without a source; read at https://www.theartstory.org/artist/miro-joan/ . Uncertain.
• [MAL25] Fundació Miró Mallorca, news item on an X-ray fluorescence survey of its collection, October 2025. Read through a summary. https://miromallorca.com/es/descodificant-miro-un-escaner-revela-els-secrets-quimics-de-la-seva-paleta-a-mallorca-cast/
• [SAL] Field's Chromatography, revised by T. W. Salter (London 1869), §293 "Ivory Black." https://www.gutenberg.org/ebooks/20915