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 more than her own face. Her portraits of relatives, friends, a physician and the people connected to patrons show a painter responding to quite different relationships: a picture might be a family gift, an act of gratitude, a memorial or a commission. Those circumstances do not turn every likeness into the same kind of transaction, and the people represented are not always securely identified. [1] [2] [5] [8] [12]

Selected works reveal the range of this practice without amounting to a complete catalogue of everyone Kahlo painted. Some sitters are named by an inscription or a museum; for others, even a detailed painting leaves the name uncertain. The difference between a documented identity and an imagined biography is part of these portraits' history. [2] [3] [4] [6]

At a Glance

  • Range: Family, friends, a physician, memorial subjects and patrons appear in distinct portrait settings. [1] [8] [9] [12] [14]
  • Identities: Some sitters are named; the Lady in White and several child portraits retain significant uncertainty. [4] [5] [6] [7]
  • Patronage: Morillo's five named family portraits in one museum account differ from his collection of about thirty Kahlo works. [14] [15]
  • Two artists: Kahlo and Rivera made separate 1943 portraits of Natasha Gelman for different members of the patron couple. [16] [17]

Contents

  1. Introduction
  2. Family Portraits Across a Lifetime
  3. Children Whose Stories Are Not Equally Known
  4. Friends and a Physician in San Francisco
  5. When the Subject Was Not Sitting
  6. Portraits for Patrons
  7. Looking Beyond a Single Meaning
  8. Related Reading

Family Portraits Across a Lifetime

In Portrait of Cristina, My Sister (1928), Kahlo presented her younger sister in a signed oil painting on panel. Christie’s records that she gave it to Cristina and that Cristina’s daughter Isolda inherited it around 1964. This was a picture with a documented family recipient, but the auction catalogue’s comparison with Renaissance models and New Objectivity is an interpretation of its form, not a statement by Kahlo about why she painted her sister. [1]

Christie’s essay also passes on a reading of Cristina as a modern Eve. Whatever that comparison contributes to looking at the 1928 picture, it cannot make the portrait a prediction of Cristina’s later relationship with Diego Rivera. The dates and the available evidence are more precise than a retrospective story in which every earlier image foretells a later family conflict. [1]

The following year, Kahlo painted Cristina’s infant daughter in Portrait of Isolda Pinedo Kahlo. The current SURA record identifies the 1929 oil canvas and its child sitter, while close study of the painting distinguishes the flower in Isolda’s hand and the rag doll beside her. Isolda’s later inheritance of the Cristina portrait belongs to that separate work’s provenance; it does not make her the subject of her mother’s picture. [1] [2]

Family portraiture was not confined to Kahlo’s youth. Portrait of My Father, catalogued as a 1952 painting, gives Guillermo Kahlo a later memorial presence, while a Gisèle Freund photograph dated 1951 shows Frida already at work on it. The words painted beneath his likeness tell us how she chose to honour him; they are not independent proof of every genealogical or political assertion the inscription contains. [3]

These three works have different recipients, dates and purposes. One was given to a sister, another depicts the sister’s child, and the later work commemorates a father after his death. The relationships matter, but none removes the need to look at the particular painting and its own documentary record. [1] [2] [3]

Kahlo also painted Diego Rivera as a separate portrait subject. Fundación Proa records her 1937 Portrait of Diego Rivera as an oil on Masonite measuring 46 by 32 centimetres, a close likeness quite unlike the miniature Rivera face later embedded in her own Diego en mi pensamiento. A spouse’s portrait can belong to an account of family and intimacy, but the relation alone does not tell us why she selected this particular format or what Rivera thought of it. [19]

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.

Children Whose Stories Are Not Equally Known

Kahlo’s 1929 La niña Virginia has a named child and a striking second life on the same support: a study for a self-portrait with an aeroplane appears inverted on its reverse. The Museo Dolores Olmedo-linked catalogue records the portrait as oil on Masonite, while its bilingual commentary describes the child’s dress, chair and direct gaze. These observations do not recover Virginia’s full biography or establish who requested the picture. [4]

Two other 1929 portraits illustrate why a descriptive title is not a sitter identification. The unfinished Portrait of a Girl in the Museo Frida Kahlo presents a seated child whose name is not established by the current museum and catalogue records. Girl with Necklace is a different oil canvas: Sotheby’s supplies its title, date, material and later provenance, but not the young sitter’s identity or the circumstances of its commission. [5] [6]

Their shared date and youthful subjects do not establish a common patron, place or social background. They are three distinct paintings, each with a different extent of surviving information. Even Virginia’s name tells us less about her life than a fuller biographical record would. [4] [5] [6]

The adult Portrait of a Lady in White, also dated around 1929, is an especially instructive comparison. Christie’s documents its larger oil canvas and early gift to the photographer Lola Álvarez Bravo, yet offers two competing sitter proposals, Dorothy Brown Fox and Elena Boder; the painted pink banderole gives no name. Its dimensions resemble those of the unfinished Portrait of a Girl, but the adult figure, different painting and separate records rule out treating the two as one object. [5] [7]

Christie’s discusses photographic and family evidence and cites a specialist IFAR study, but its account leaves the identity open. Brown Fox and Boder remain proposals, not alternative spellings for a securely named sitter. That unresolved name is a significant difference between this portrait and works whose inscriptions identify the person shown. [7]

Friends and a Physician in San Francisco

Kahlo’s Portrait of Eva Frederick is dated 1931, during the period she spent in San Francisco. A painted inscription gives Eva’s name and New York birthplace, and the Museo Dolores Olmedo-supplied commentary calls her an African American friend. Against the picture’s close frontal presentation, her dark dress, necklace and hair arrangement are visible; her age, profession, introduction to Kahlo and payment or sitting terms are not securely documented. [8]

The uncertainty matters because the portrait has attracted readings about dignity, race and identification. Josefina García’s catalogue essay offers an interpretation of Kahlo’s formal choices, but a strong social meaning should be attributed to the writer rather than presented as Kahlo’s own recorded motive. A named sitter is not the same as a fully recoverable life. [8]

The same museum also records a separate 1931 pencil-and-charcoal Nude of Eva Frederick. Its different medium and pose show that Eva was not represented only in the oil portrait, but neither object establishes the sequence of the two works or proves that the drawing was a preparatory study. The Madrid exhibition catalogue placed them alongside one another, a documented later curatorial pairing rather than evidence that Kahlo originally conceived a pair. [8] [20]

Dr Leo Eloesser had a more fully documented relationship with the painter. Zuckerberg San Francisco General says Kahlo made his 1931 full-length portrait in his Leavenworth Street home, as a gesture of friendship and repayment for medical care. The physician stands beside a table and model boat, with a small image of Rivera also in the room: these objects give the likeness a setting unlike Eva Frederick’s compressed portrait. [9]

Maureen and Robert Park, writing about patient–doctor relationships in art, read the Eloesser portrait as gratitude and compare it with Kahlo’s separate 1940 Self-Portrait Dedicated to Dr Eloesser. Their description of his pose as awkward or aloof is a critic’s reading of the image, not evidence of the man’s character. Nor does the general fact that artists sometimes paid doctors with paintings establish a specific fee or invoice for this work. [9] [10]

The two San Francisco portraits therefore resist a single “friends” formula. Eva is identifiable but biographically elusive; Eloesser’s link to care is documented and the scene surrounds him with objects. The differences in what Kahlo painted and what later records preserve matter more than guesses about either sitter’s inner life. [8] [9] [10]

When the Subject Was Not Sitting

Portrait of Luther Burbank (1931) shows how far Kahlo could move from a conventional live likeness. The American plant breeder had died in 1926; Kahlo joined his human figure to a trunk and roots in a posthumous image. The Museo Dolores Olmedo-linked exhibition catalogue identifies the work, while its plant-and-body interpretation belongs to curatorial reading rather than a universal code Kahlo left for every root. [11]

No contemporary commission or sitting can be inferred from the painting’s title. Its difference from Eloesser’s portrait lies partly in that historical fact: one relationship could be documented through living acquaintance and medical care, while a posthumous public figure was made present through a transformed body. The portrait category includes both without pretending they were produced in the same way. [9] [11]

The Suicide of Dorothy Hale offers another difficult boundary. Clare Boothe Luce commissioned the work in 1938, and Kahlo developed it through the end of that year and into 1939 as a narrative memorial of Hale’s death rather than a conventional living-sitter portrait. Phoenix Art Museum owns the oil-on-Masonite painting with its painted frame, while Tate’s provenance and Kahlo’s dated letters supply a firmer sequence than a single date on a label can convey. [12]

Art historian Fiona Gregory shows that later accounts differ over who proposed the commission, what kind of image Luce expected and for whom it was intended. The picture’s inscription was altered, but the responsible person and exact intervention cannot simply be declared settled. For a wider survey of portraits, the essential point is that a commissioned likeness could become a disputed memorial exchange; Hale’s life and the painting’s ethical history require their own fuller account. [12] [13]

Portraits for Patrons

Eduardo Morillo Safa and the Gelmans were patrons, but their documented arrangements were not identical. A bilingual Museo Dolores Olmedo-supplied account says Morillo commissioned portraits of five named family members—his mother Rosita, wife Alicia, daughter Lupita, son Eduardo and himself—and collected about thirty works by Kahlo. Thirty is a collection figure, not the number of family portraits he ordered. [14]

In the 1944 Portrait of Doña Rosita Morillo, his mother sits knitting before a dense plant background. The current owner-linked catalogue gives oil on Masonite and describes the portrait as more natural in its treatment than the related engineer likeness; other meanings proposed for Rosita’s work or surrounding cactus belong to their authors. A separate thesis includes another daughter, Mariana, in a six-person family count, leaving the exact extent of the family series unsettled. [14] [15]

The documented family commission was not a Mexican government order, and Eduardo’s later diplomatic appointments should not be projected back onto the making of the 1944 picture. His relationship with Kahlo as collector and patron is substantial without a government programme or a thirty-portrait series. [14] [15]

For Natasha Gelman, institutional accounts describe a two-artist arrangement in 1943. Natasha commissioned Kahlo’s small upright head-and-shoulders portrait, while Jacques Gelman commissioned Diego Rivera’s much larger reclining portrait of Natasha amid flowers. The contrast shows two different pictorial responses within the same patrons’ circle; it does not document a rivalry, exact order of sittings or Kahlo’s private judgement of Rivera’s painting. [16] [17]

The pair must also remain separate in current records. Santander’s 2026 announcement says it will manage the Gelman collection, not that the bank bought Kahlo’s individual portrait; its press photograph labelled Portrait of Natasha Gelman shows Rivera’s landscape canvas. Kahlo’s smaller image has a separate object history, including a published disagreement over whether its support is Masonite or canvas, which should not be settled by a picture of Rivera’s work. [16] [17] [18]

Looking Beyond a Single Meaning

Kahlo’s portraits of other people range from close family likenesses to the markedly constructed bodies of Burbank and Hale. A painted name, a gift record, an owner’s label and a critic’s symbolism answer different questions; none can stand in for all the others. This is why the unknown child sitters and disputed Lady in White identity belong beside the better documented Morillo and Eloesser relationships. [1] [4] [5] [7] [9] [11] [12] [14]

The selected works do not add up to a complete count of Kahlo’s portraits. They do show a practice more varied than the familiar emphasis on self-portraiture: Kahlo used likeness to address family, acknowledge care, transform a public figure, negotiate remembrance and work with patrons. Each purpose shaped a different picture, and the visible differences do more than illustrate her biography. [1] [8] [9] [11] [12] [14] [16]

Portrait of Eva Frederick and Portrait of Dr. Leo Eloesser develop two different San Francisco relationships. The Suicide of Dorothy Hale examines the contested memorial commission; Portrait of Doña Rosita Morillo and Portrait of Natasha Gelman follow two patronage settings. The self-portraits guide considers Kahlo as her own sitter.

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

How does a named sitter differ from a person whose identity remains a proposal?

Context

What changes between the familiar person, the posthumous figure and the memorial subject?

Open Question

How do the Morillo and Gelman arrangements show different roles for patrons?

Join the Conversation

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

References

  1. Christie’s, Portrait of Cristina, My Sister (1928), lot details, provenance and essays, especially lot lines 11–24 and 67–77. The essay transmits selected Lozano readings from the catalogue, not its complete argument. View source text.
  2. Arte y Cultura SURA, Retrato de Isolda Pinedo (1929), current Spanish object record; The painting shows a flower and rag doll; earlier published sizes differ. View source text.
  3. Museo Frida Kahlo, La Casa Azul: el universo íntimo de Frida Kahlo, p. 1, naming Retrato de mi padre; museum-supplied Gisèle Freund photograph record dated 1951 documents work in progress. View museum guide; view photograph record. View source text 1; View source text 2.
  4. Fundación Casa de México en España / Museo Dolores Olmedo, La niña Virginia bilingual object commentary and 2022 catalogue, printed pp. 21–23 and 60, including the reverse study. View source text.
  5. Museo Frida Kahlo-supplied Retrato de Muchacha (inconcluso) object record and Museo Roca catalogue entry 035, printed p. 73, identify the separate unfinished 1929 child canvas. View museum object text. View source text.
  6. Sotheby’s, Niña con collar (1929), New York sale N09579, lot 17, object and provenance details; no sitter identification is supplied. View source text.
  7. Christie’s, Portrait of a Lady in White (circa 1929), object and provenance lines 9–19 and competing sitter discussions lines 60–88. The listed 2013 IFAR article has not been consulted in full. View source text.
  8. Fundación Casa de México en España / Museo Dolores Olmedo, Retrato de Eva Frederick bilingual object text; Josefina García’s Spanish catalogue essay and checklist in Frida Kahlo: alas para volar, printed pp. 21 and 60. View source text.
  9. Zuckerberg San Francisco General, “Portrait of Dr. Leo Eloesser and La Tortillera,” owner account of making, relationship, object and later donation. Its old current-loan line is not used to claim present display. View source text.
  10. M. P. Park and R. H. R. Park, “The fine art of patient-doctor relationships,” BMJ 329 (2004), pp. 1475–80, especially p. 1478 on Kahlo and p. 1479 conclusion. Their clinical biography is secondary and their pose language attributed. View source text.
  11. Fundación Casa de México en España / Museo Dolores Olmedo, Frida Kahlo: alas para volar (2022), printed pp. 34–36 and 60, Portrait of Luther Burbank account; A related drawing is a separate object. View institutional catalogue. View source text.
  12. Phoenix Art Museum, The Suicide of Dorothy Hale owner line; Tate, Frida: The Making of an Icon object/provenance entry X102248; Two dated Kahlo letters establish work continuing from December 1938 into January 1939. View owner text; view Tate record. View source text 1; View source text 2.
  13. Fiona Gregory, “Death, Fashion and Feeling: Reading around The Suicide of Dorothy Hale (1939),” Affirmations: of the Modern 1, no. 2 (2014), pp. 39–55, especially pp. 39–43 and 47–49. The complete publisher text was reopened; cited earlier recollections remain mediated. View source text.
  14. Fundación Casa de México en España / Museo Dolores Olmedo, bilingual Portrait of the Engineer Eduardo Morillo Safa text, lines 71–78, distinguishing five named family portraits from approximately thirty collected works. View source text.
  15. Fundación Casa de México en España / Museo Dolores Olmedo, Frida Kahlo: alas para volar (2022), printed pp. 33–34 and 60, Portrait of Doña Rosita Morillo, owner metadata and attributed formal reading; A UNAM thesis instead lists six family members. View institutional catalogue. View source text.
  16. Metropolitan Museum Gelman collector account and Fundación Proa commission/collection history ; the two patrons commissioned distinct Kahlo and Rivera images. View Proa collection history. View source text.
  17. Frist Art Museum, “Frida Kahlo, Diego Rivera, and Mexican Modernism from the Jacques and Natasha Gelman Collection,” institutional account of Jacques’s Rivera commission and distinct large painting, lines 59–62 and 90–94. View source text.
  18. Santander, January 2026 Spanish Gelman collection-management announcement, lines 62–70, 114–26 and 366–68. The press plate is Rivera’s work, not Kahlo’s. View source text.
  19. Fundación Proa, Retrato de Diego Rivera (1937), Spanish object entry giving oil on Masonite and 46 × 32 cm; this is distinct from Rivera images inside Kahlo's self-portraits. View source text.
  20. Museo Dolores Olmedo-supplied record, Nude of Eva Frederick (1931), pencil and charcoal on paper, distinct from the oil portrait; 2022 Casa de México catalogue physical pp. 28–29 and checklist p. 64 juxtaposes the works. View drawing record; view exhibition catalogue. View source text 1; View source text 2.