by
Tom Gurney BSc (Hons) is an art history expert with over 20 years experience
Page created on March 2, 2018 / Page updated on October 6, 2026
Email: tomgurney1@gmail.com / Phone: +44 7429 011000

Introduction

Frida Kahlo painted on canvas, metal, wood and manufactured boards, made a portable fresco, and worked on paper with ink, pencil and watercolour. [1] [2] [3] [4] These are documented examples, not a statistical map of every work she made. The best technical studies also show that her procedure could combine an initial design with substantial changes: underdrawing, revised forms, altered colour and even a covered inscription survive in examined objects. [5] [6]

No single recipe explains all of these works. A museum's medium line identifies an object at catalogue level; infrared imaging or X-radiography can reveal an earlier state; a microscopic sample can identify material only where it was taken; and a collaborator's recollection is testimony rather than a laboratory result. This article moves among those kinds of evidence without turning a finding from The Two Fridas or three Dallas paintings into a claim about Kahlo's entire oeuvre. [5] [6] [7]

At a Glance

  • Scope: Documented examples of canvas, metal, wood, rigid boards, fresco and works on paper—not a statistical census of Kahlo's oeuvre. [1] [2] [3] [4]
  • Method: An owner's medium line, imaging study, microscopic sample and recollection answer different questions and must not be merged. [5] [6] [7]
  • Painted components: Centre Pompidou catalogues The Frame as oil on aluminium within reverse-painted glass and a painted wooden frame. [3]
  • Scope of findings: Findings from The Two Fridas and the Dallas studies concern those examined objects; they do not describe every Kahlo painting. [5] [6]

Contents

  1. Introduction
  2. What Counts as Material Evidence
  3. Canvas and the Layers Beneath It
  4. Metal Is Not One Material
  5. Rigid Boards, Wood and a Portable Fresco
  6. Paper, Drawing and the Diary
  7. Preparation, Underdrawing and Revision
  8. Brushwork and the Artist's Hand
  9. Pigments, Whites and Deterioration

What Counts as Material Evidence

The first question is what the paint or drawing sits on. “Oil on canvas” names a paint medium and support but says little about fibre, ground, weave, varnish or treatment. “Oil on metal” conceals whether an institution has identified zinc, aluminium, tin-plated sheet or only an undetermined plate. “Masonite” is a branded hardboard term that museums and older catalogues may use alongside hardboard, fibreboard and composition board; those labels should not be silently made identical. [2] [3] [4] [8]

Technical evidence answers narrower questions than a familiar phrase such as “Kahlo's palette.” Imaging can disclose shapes beneath a surface; examination under magnification can distinguish paint handling; chemical methods can identify or propose compounds in a local sample. A screen image can show apparent colour and arrangement, but not the pigment, binding medium, unseen layers, precise texture or condition of the original. [5] [6] [7] The most useful rule is to name both the method and the particular object whenever making a technical claim.

Watch: Frida Kahlo

Five perspectives to view here alongside the article.

Frida Kahlo’s The Two Fridas: Great Art Explained

Great Art Explained

A focused visual introduction to the painting and its context.

Canvas and the Layers Beneath It

MoMA catalogues Self-Portrait with Cropped Hair as oil on canvas, 40 × 27.9 centimetres. [1] That reliable owner line is not a conservation report: the cited public record does not identify its weave, sizing, ground, pigments, varnish or treatment history. Those details cannot be borrowed from another Kahlo canvas simply because its catalogue medium is similar.

The Two Fridas is also oil on canvas, but it has been studied far more closely. A 2025 peer-reviewed article, using earlier imaging and sampling alongside later microscopy, describes a medium-weave cotton canvas, possible light hide-glue sizing and a red preparatory-looking layer in one sampled area. [5] The authors do not find a uniform conventional ground across the examined passages. That does not prove Kahlo never grounded any canvas, nor that every part of this painting has exactly the same layer structure. [7]

Some Kahlo works combine canvas with a rigid backing. A catalogue may call such an object “canvas on Masonite” or another fibreboard, which is different from oil applied directly to the board. [4] Support, backing, paint layer and frame need to be distinguished before an object is used to argue for a particular technique. Even apparently minor wording differences can change the physical history a reader imagines.

Metal Is Not One Material

Kahlo's small metal-supported works are often discussed in relation to Mexican retablos and ex-votos. Teresa del Conde's Spanish study argues for a deliberate engagement with popular painting traditions and notes sheet-metal formats. [9] That historical argument does not establish the alloy, primer or workshop preparation of every individual plate. The broader religious and popular-art history belongs with the retablo study; here the material distinction matters.

MoMA calls My Grandparents, My Parents, and I oil and tempera on zinc. [10] Zinc is not a synonym for tin, and the owner metadata does not by itself prove a specific plate manufacturer or paint-layer sequence. Centre Pompidou is equally specific about the painted self-portrait in The Frame: oil on a thin aluminium plate. [3] The aluminium sits within reverse-painted glass from Juquila and painted wood, so a single phrase such as “oil on glass” would misidentify the surface on which the portrait was painted. [11]

Elsewhere, object records use “tin plate,” “metal” or competing descriptions. Those should be reported with their source and uncertainty rather than forced into an invented uniform category. [4] In The Bus, for example, recent owner-linked material records canvas while older published captions give metal; without an item-level technical file, the conflict remains a conflict. [12] Material precision is not pedantry when the possible supports are physically different.

Rigid Boards, Wood and a Portable Fresco

The three works examined in Dallas—Sun and Life, Diego and Frida 1929–1944 and Still Life with Parrot and Flag—are all described there as oil on Masonite. [6] Masonite is a type of hardboard, but that owner-linked term does not answer whether each board was primed, what fibres it contains, whether a backing was added or how it has aged. The cited records do not provide a complete technical survey of Kahlo's boards. Their availability, portability, surface and relation to popular formats are questions worth testing, not documented reasons for every choice. [4] [9]

Christie's records Portrait of Cristina, My Sister as oil on panel. The composition extends beyond its central wooden panel into the surrounding stretcher or frame structure, making the object's edge part of the painted design. [14] Abby McEwen's catalogue essay compares the work with Renaissance portrait traditions and relays Luis-Martín Lozano's account of Kahlo learning photographic retouching in Guillermo Kahlo's studio. Those are historical interpretations and mediated testimony, not wood-species identification or proof that a photographic process was mechanically transferred into this painting.

An even more unusual case is Self-Portrait (Very Ugly) of 1933. Christie's current record calls it fresco on Celotex board, 27.3 × 22.2 centimetres; an earlier Dallas account called the portable support plasterboard. [15] [16] The two terms should be shown as source-specific descriptions rather than blended into a fictitious material. Dallas places the experiment within the context of portable plaster panels and Mexican mural practice, but this one work cannot establish how long Kahlo pursued fresco or how she prepared every later surface.

Lucienne Bloch recalled that she and Stephen Dimitroff made a first panel that dried before Kahlo could paint; they prepared a second, which became the surviving work. [15] This is a rare account of making as a timed, collaborative practical process, though it reaches us through later publication and auction text rather than a freshly read original note. Bloch remembered a fuller inscription than the abbreviated words now on the object. Dallas describes damage and cites a historical photograph and Bloch's colour copy as evidence of an earlier appearance; those witnesses do not authorise us to reconstruct or reproduce a supposedly pristine original. [15] [16]

Paper, Drawing and the Diary

Kahlo's material practice did not end at oil painting. El Salón de Belleza is catalogued as watercolour, pen and black ink and pencil on paper; it is a work on paper, not another example of oil on rigid support. [17] The distinction matters when speaking of “paintings,” because a coloured drawing and an oil canvas can create related imagery while requiring different making and conservation methods.

French scholarship on the diary describes writing, lists, drawing, ink and colour working together on pages. [18] [19] Hourdin and Suárez caution that few finished paintings can be identified there in germ, weakening a simple sketchbook-to-canvas narrative. [18] Noël and Oberhuber's author-hosted study draws attention to transfer between pages, overwriting and changes in colour, but its publication venue and date remain unconfirmed. [19] These sources illuminate paper as a worked material object, not a transparent record of every private thought or a universal preparatory notebook.

A different paper object brings technical evidence into the discussion. In a 2017 Museo Soumaya publication, conservator Rebeca Rosendo Corona reports on a 2005 examination of two paintings attributed to Kahlo on the front and back covers of a Pierre Cazeaux obstetrics book: Frida como calavera y Frida ante el espejo and Diego como calavera. [28] The publication is a museum account of a report retained in Raquel Tibol's archive, not the complete signed laboratory report itself. Its unusual book-cover support expands the documented material comparison, while the precise object identity and authentication case remain tied to that mediated source.

The same boundary applies to depicted materials. A painted blouse represents textile, but its visible embroidery is made from paint; the V&A's account of huipil weaving and stitching answers a garment-history question, not a pigment-analysis question. [20] [21] One source cannot substitute for the other. The full histories of Kahlo's clothing and graphic work belong in their specialist pages rather than being used to inflate a painting-technique survey. [20] [22]

Preparation, Underdrawing and Revision

Dallas Museum of Art conservator Laura Eva Hartman examined Still Life with Parrot and Flag, Sun and Life and Diego and Frida 1929–1944 with imaging and close surface study. [6] Infrared photography of the still life revealed an initial drawing made with both thin lines and broad ink strokes. Fruit forms changed in size and shape, while work around the parrot's wing, beak and adjacent mango was particularly active. [6] [23] The image was planned, but not fixed at its first drawn state.

X-radiography of Sun and Life revealed an even more consequential sequence. Several forms around the sun began as closed pods and opened during painting; a fetus-like element behind the sun appeared late where an earlier closed pod had been. [6] [13] That finding tells us a visible motif developed through revision. It does not tell us why Kahlo made the change, whether she described it as a fetus, or how every related painting was built.

In The Two Fridas, technical study found that the viewer-right heart moved towards the centre and that an earlier purple state lies beneath part of the present viewer-left white skirt. [5] Those observations challenge the idea of a final composition simply transferred in one step. They also limit an interpretation that treats the current white dress as the only colour Kahlo ever considered for that passage. The sequence and sampling are specific to this canvas; no general law of her working order follows. [7]

The Dallas group includes a different kind of revision. Infrared study of a shell on the frame of Diego and Frida 1929–1944 uncovered “Recuerdo de Veracruz” beneath red paint. [6] Hartman says Kahlo probably covered the phrase, a probability worth retaining rather than upgrading to certainty. The inscription shows that a frame could carry and later conceal information, but it does not by itself document a commission, purchase in Veracruz or the motive for repainting.

Together the examined works support a careful general statement: Kahlo could establish a strong compositional direction and still adjust drawing, shapes, colour, anatomical position or text. [5] [6] They do not prove that she always drew first, always revised late, or followed one sequence from preliminary image to finished canvas. For works without an object-specific X-ray, infrared record or conservation file, those questions remain open.

Brushwork and the Artist's Hand

Hartman describes different paint handling across the three Dallas works. Still Life with Parrot and Flag builds textures through small rapid strokes; Sun and Life includes longer sinuous movements; Diego and Frida 1929–1944 uses short, tiny strokes in a more direct, thickly textured application. [6] Finger impressions survive in paint on two of the examined objects. These details complicate a stock description of Kahlo as always uniformly smooth and miniaturist.

Precise representation need not mean identical handling. A fruit, leaf, face, shell or ground plane may be treated differently within and between works, and magnification can expose the movements that an ordinary reproduction conceals. [6] The accessible sample of Jodi Roberts's MoMA book on Self-Portrait with Cropped Hair analyses palette and animated hair as a visual construction, but it is not a conservation report and does not identify that canvas's pigments or layer order. [25] It should not be used to fill the MoMA object's technical gaps.

Kahlo's early contact with Guillermo Kahlo's photographic studio is relevant to skill and attention. McEwen relays Lozano's discussion of retouching, composition and light-and-shadow lessons, but the exact influence on Portrait of Cristina is an art-historical proposition rather than a demonstrated transfer process. [14] The technical record lets us replace the false opposition between “self-taught spontaneity” and “mechanical precision” with work-specific observation.

Pigments, Whites and Deterioration

The Two Fridas study and its supporting information provide the most detailed pigment evidence cited here, and even they concern sampled passages of one painting. [5] Analyses identify zinc- and lead-containing whites, reds including vermilion and cadmium or iron-based materials, and yellow passages with chrome yellow and ochre; some blue identifications are proposed as possibilities rather than certainties. Mixtures of pigments with similar apparent colours helped build nuanced tones. This is an examined palette in particular locations, not a list of Kahlo's lifelong favourites.

The earlier purple skirt layer may have involved an organic colourant, but the study does not identify a commercial tube or brand that Kahlo definitely used. [5] A historical tube comparison in the supplement is a research aid, not proof of her studio stock. Likewise, an apparently white area in a museum JPEG cannot be called “zinc white” without material examination. A colour name describes sight; a pigment name is an analytical claim.

The Soumaya account reports a different local analysis. For the two book-cover paintings, the 2005 examiners used magnification, ultraviolet light, microchemical testing and microscopic layer examination; they reported drying oil, surface varnish and no preparatory ground or later additions in their samples. [28] The museum's published list includes red oxide, synthetic red lake, vermilion, ultramarine and Prussian blues, zinc and barium whites, gypsum, cadmium yellow, ochre, umber and carbon black. That list belongs to the examined covers as reported by Rosendo Corona, not to every Kahlo work; the underlying laboratory report and sampling map have not been read.

The same article investigates cracking and other change in white passages associated with zinc white, including zinc lactate formation. [5] Its proposed mechanism considers oil, moisture and the way paint interacts with the canvas system; the authors retain qualifications rather than presenting one complete account of every crack. Possible limited sizing and the absence of a continuous conventional ground in sampled areas may matter, but they do not establish that Kahlo never prepared her canvases. [7]

DDT was also detected in the examined painting. The researchers found no documentation of who applied it, when or why, so detection cannot be turned into a story about Kahlo's own studio practice, health or a known treatment. [5] This is a useful reminder that a chemical trace and its historical cause are different questions. Conservation studies often become most informative when their negative findings are stated as carefully as their positive ones.

What Technical Methods Can and Cannot Show

Different tools answer different questions; none reads intention directly. Infrared imaging may reveal some underdrawing or a covered inscription when materials respond to the equipment, as in the Dallas works. X-radiography maps differences in X-ray absorption and can expose structural changes such as the earlier Sun and Life pods. Ultraviolet examination may reveal fluorescence useful for studying coatings or alterations, but no claim in this article rests on a Kahlo UV image alone. [6] [23]

Microscopy examines surface, fibres and tiny cross-sections at a different scale. Scanning electron microscopy with elemental analysis can map elements; infrared spectroscopy and other molecular techniques can help characterise compounds; chromatographic methods can address some organic materials. [5] [24] An element map, fibre observation and molecular identification are complementary, not interchangeable. The Two Fridas team used multiple methods precisely because a single reading could not answer every layer, pigment and degradation question.

General museum method pages explain how these tools work, but they are not evidence that Kahlo herself used a named material or that a particular painting received treatment. [23] [24] Treatment choices require condition, risk, ethics and object history in addition to analysis. The cited sources do not include complete owner conservation files for most of these works, so their present condition and treatment needs cannot be established here.

Frames Are Part of Material History

The frame around The Frame is not a neutral border. Centre Pompidou catalogues an aluminium portrait, Juquila reverse-painted glass and painted wood as components of one work. [3] The viewer sees colour through the glass to paint on its reverse, while the central portrait has a different support and vulnerability. The public object record does not disclose a complete condition or treatment history for either part. [11]

The shell frame of Diego and Frida 1929–1944 bears the covered Veracruz phrase, and Portrait of Cristina extends its painting beyond an inserted panel. [6] [14] In each case, a frame carries evidence of making or alteration, not merely presentation. Mirrors and glass also change a viewer's experience, but the detailed account of reflection and self-portraiture belongs elsewhere; the technical point here is that an assemblage cannot be reduced to the medium of its central image.

The Two Fridas study discusses findings from examination of that particular canvas, while The Frame distinguishes its painted metal, glass and wooden components. Drawings follows Kahlo's work on paper, and How to Read Frida Kahlo explains how to keep observation, catalogue facts and interpretation apart.

Further Reading and Open Questions

TASCHEN's Frida Kahlo: The Complete Paintings offers object-by-object entries, medium terminology and bibliographies; Christie's citation of individual entries does not establish every detail in the complete volume. [26] Jodi Roberts's MoMA study focuses on Self-Portrait with Cropped Hair, with an official sample available for direct reading. [25] Neither book supplies a technical conclusion about works outside the passages actually cited here.

Full conservation files for the Dallas works, The Two Fridas, The Frame and MoMA's cropped-hair portrait could clarify layer structure, alterations and treatment history beyond the available summaries. [7] The 2005 Serrano Cabrera–Vázquez Negrete report behind Soumaya's account, Bloch's original recollection and the source underlying Lozano's retouching account likewise require their own exact texts. [28] Any new finding would apply first to its examined object, not automatically to every Kahlo painting.

A technical study's X-ray, micrograph or figure is a distinct visual record, often with separate credits and rights. [27] Its findings may illuminate one examined work without making the figure a freely reusable image or a general account of Kahlo's materials.

Discussion

Use the cited evidence and your own close looking to explain your view. The prompts below are starting points, not fixed interpretations.

Reader Insights

Close Looking

What can a museum medium line establish—and what can it not?

Context

How does technical imaging change a first visual reading?

Open Question

Why should a finding from one work not become an oeuvre-wide rule?

Join the Conversation

Which reading best fits the evidence, and where does uncertainty remain?

References

  1. The Museum of Modern Art, Self-Portrait with Cropped Hair, current owner medium and size record. View source text.
  2. The support examples are drawn from MoMA's canvas and zinc object records (references 1 and 10), Centre Pompidou's component record (reference 3), and the Dallas conservation account (reference 6); these records do not constitute an oeuvre-wide materials survey.
  3. Centre Pompidou, The Frame / Le cadre, component record for oil on aluminium, Juquila reverse glass and painted wood. View source text.
  4. The varied supports named here are documented by the owner and object records in references 1, 3, 6, 10, 14 and 15. This selection is illustrative, not a statistical census of all authenticated Kahlo works.
  5. Pablo Aguilar-Rodríguez and colleagues, technical study of Las Dos Fridas, ACS Omega 10 (2025), complete article and twelve-page supplement read; local sample and uncertainty limits retained. View source text.
  6. Laura Eva Hartman, “Conserving Kahlo,” Dallas Museum of Art, 29 March 2021, complete public account of three examined works, imaging, revisions, handling and shell frame. View source text.
  7. Aguilar-Rodríguez and colleagues' study (reference 5) concerns sampled passages of The Two Fridas; Hartman's Dallas account (reference 6) concerns three other examined objects. Neither establishes a technique for Kahlo's entire oeuvre.
  8. The medium lines in the owner records cited in references 1, 3, 6 and 10 name particular objects and components. No percentage distribution across Kahlo's entire output follows from those examples.
  9. Teresa del Conde, “Lo popular en la pintura de Frida Kahlo,” Anales del Instituto de Investigaciones Estéticas 13, no. 45 (1976), 195–203, complete article read for historical format argument, not plate-alloy identification. View source text.
  10. The Museum of Modern Art, My Grandparents, My Parents, and I, current oil-and-tempera-on-zinc owner record. View source text.
  11. Centre Pompidou's object record identifies the aluminium portrait, Juquila reverse-painted glass and painted wood; its public component description is not a complete condition or treatment report. View source text.
  12. Mexico's Secretaría de Cultura item credit gives El camión as oil on canvas, 26 × 55.5 cm; Andrea Kettenmann's Frida Kahlo 1907–1954: Dolor y pasión (1999 Spanish edition), plate caption, gives oil on metal and 25.8 × 55.5 cm. No technical file has resolved the published support conflict. View source text.
  13. Dallas Museum of Art conservator Laura Eva Hartman's account describes X-radiography of Sun and Life and the changes from closed to opened forms. It is a public conservation summary, not the full examination file. View source text.
  14. Christie's / Abby McEwen, Portrait of Cristina, My Sister, lot 6453089, complete catalogue record and essay read; panel and photographic influence kept at different evidence levels. View source text.
  15. Christie's / Abby McEwen, Self-Portrait (Very Ugly), lot 6397640, current Celotex line and mediated Bloch account. View source text.
  16. Erin Piñon, “Frida in Fresco,” Dallas Museum of Art, 13 June 2016, earlier plasterboard terminology, historical photograph and copy. View source text.
  17. Christie's object record for El Salón de Belleza, Details section, catalogues watercolour, pen and black ink and pencil on paper, rather than oil on a rigid support. View source text.
  18. Gaëlle Hourdin and Modesta Suárez, “Le Journal de Frida Kahlo: des pleins et des liés”, Mélusine XXXIII (2013), printed pp.183–193 / PDF pp.183–193. Its diary-material analysis is distinct from the technical examination of an oil painting. View source text.
  19. Marine Noël and Andrea Oberhuber, “Le Journal de Frida Kahlo: la matérialité de l’objet livre et son lien au vivant”, author-hosted 15-page manuscript, sections on the diary's materials and page design. A final publication venue and date have not been verified. View source text.
  20. Chloë Sayer, “Traditional Mexican dress,” Victoria and Albert Museum, complete textile article, not paint analysis. View source text.
  21. For the garment history, see the companion Fashion and Tehuana Dress article and Chloë Sayer's V&A textile account in reference 20; neither is a pigment-analysis source.
  22. For the broader graphic-work context, see the companion Drawings article and the diary-material studies in references 18 and 19. These do not turn a diary leaf into a technical study of an oil painting.
  23. Museo Nacional Thyssen-Bornemisza, “Technical Images”, general explanation of imaging methods; no Kahlo-specific result. View source text.
  24. Museo Nacional Thyssen-Bornemisza, materials laboratory, general analytical-method reference; no Kahlo-specific treatment evidence. View source text.
  25. Jodi Roberts, Frida Kahlo: Self-Portrait with Cropped Hair official sample, MoMA One on One, 12-page publisher preview. The link does not represent access to the complete book. View source text.
  26. TASCHEN, Frida Kahlo: The Complete Paintings, publisher record only; complete 624-page volume unread. View source text.
  27. The technical figures discussed in the cited conservation and imaging studies have their own image credits and rights. Access to a study does not itself grant permission to republish those separately credited figures.
  28. Rebeca Rosendo Corona, “Frida y Diego como calaveras, Archivo Raquel Tibol,” in Museo Soumaya, Frida Kahlo según Raquel Tibol (July 2017), p. 36, complete one-page museum account read; underlying 2005 technical report unread. View source text.