Completed British 19th C, except portraits, Portraits: British 19th C 23 Who modelled for Frederick Sandys’ painting 'Queen Eleanor'?

Queen Eleanor
Topic: Subject or sitter

Could it be Keomi Gray?

Anton Kukhto, Entry reviewed by Art UK

Completed, Outcome

This discussion is now closed. It has not been possible to identify a named model, who is considered to be the same model in a number of other paintings by Sandys.

Thank you to everyone who contributed to the discussion. To anyone viewing this discussion for the first time, please see below for all the comments that led to this conclusion.


Scott Thomas Buckle,

Betty Elzea's catalogue raisonne of Sandys' work (p334) states that Keomi must have been a Norfolk gipsy and that she makes appearances in Sandys' pictures from 1862-66. The painting of 'Queen Eleanor' is dated 1858 and the canvas is stamped 'Watson Artists' Colourman, Norwich'. An undated study in oils for the queen's head (Elzea 1.A.160) was also sourced from the same artists' supplier, suggesting that both versions were painted in Norwich. Other works by Sandys that Keomi is thought to have modelled for show a young woman with a long straight nose, brown eyes, a dark brow, and long curly hair. It is certainly very possible that Keomi modelled for Sandys in Norfolk before the stated dates and she does bear a strong resemblance to the model for 'Queen Eleanor', at least in some of the pictures identified by Elzea. In Sandys' 'Vivien', however, her chin is more pronounced (although this may be due to the low viewpoint) and in 'Judith' her features are much more mature, although this may have been painted a good few years later.

2 attachments
Betty Elzea,

I am sorry to say that I do not know who this model is. It is not Keomi, the gypsy, however.

Scott Thomas Buckle,

Different websites give different birth and death dates for Keomi, however her baptism is recorded in the Romany Traveller Family History Society baptism/birth index ( ) as follows:

GRAY Keomi
Bp 21 March 1841, St Peter, Wisbech, Cambridgeshire, daughter of Ozray and Eliza Gray, of Horsefair, Tinker.

This would make her around seventeen years of age when Sandys painted 'Queen Eleanor'.

Jacinto Regalado,

There may not have been a specific model. The face looks fairly generic to me, a more or less standard early Victorian female face. It looks more like an illustration than a portrait.

Louis Musgrove,

I have a feeling that the lady with the big chin (Vivien) might possibly be Mary Emma Jones, who was a freind of Sandy's sister.
In looking at Rosetti's paintings that are said to be Keomi, that lady has a triangular, more ethnic type of face.
Trouble is that all these pre Raphaelite ladies look a bit alike and the attributions might be a bit mixed up ! Eg. Elizabeth Siddal.

Scott Thomas Buckle,

There are certainly similarities between the facial types of models employed by Sandys. Mary Emma Jones had a long straight nose and a prominent chin, but most depictions of her tend to show her with red hair and blue eyes. Having said that, details of 'Vivien' (see attached) show a model with grey eyes and a tint of auburn in her eyebrows and hair, unlike other works that are thought to depict Keomi, where she has brown eyes and black hair and eyebrows.

In her catalogue raisonne, Betty Elzea asserts that "The model for Vivien was undoubtedly Keomi" (2.A.60., p174), but I wonder too, now, if a different model was employed, as to me as Vivien's features look quite different from the model in Sandys' 'Judith' and Rossetti's 'The Beloved'.

Louis Musgrove,

As our sitter isn't quite in the style of the usual pre Raphaelite model I wonder if it might possibly be Emma Sandy's,Frederick's sister ????

Mark Wilson,

If it's Emma Sandys she would have been 15 when this picture was painted, which could be plausible. A portrait of her by her brother done a few years previously:

when she would have been 11 or 12 shows some similarities, particularly in the colouring and the shape of the chin, though the eyebrows and nose differ somewhat and there less resemblence to a photo of the older Emma:

Louis Musgrove,

Hadn't seen the Fitzwilliam before,Emma at 11--- Yes ,a distinct possibility our sitter is Emma at 15.

I fear that it is going to be difficult to answer the question as to who modelled for this picture beyond coming up with a shortlist, or long list, of possibilities. As Louis says, "all these pre Raphaelite ladies look a bit alike". As Jacinto says, "There may not have been a specific model. The face looks fairly generic to me, a more or less standard early Victorian female face."

Martin Hopkinson,

Betty Elzea is completing a biography of Sandys and might have a view

Martin Hopkinson,

Betty thinks that without documentation this is a question which will never be answered

Betty Elzea,

I see the same model in Queen Eleanor, 1858 (Cardiff), Mary Magdalene c.1859 (Delaware), Oriana, 1861 (Tate), King Pelles Daughter, 1861 (Lanigan and Tate). There may be other pictures which show her which haven't yet 'come to light'. I have not found any clues as to who she is, but my feeling is that she was a Norwich girl. The likeness of one (unknown) person can be seen listed above, but she is not any of the other ladies mentioned by these contributors. It is a likeness not a generalisation.

I suspect that we have carried this discussion as far as possible. Unless there is new evidence, I propose to recommend closure in a week's time.

Thank you all for your input, I would be inclined to agree that there will be no more fruitful discussion and, unless new evidence is presented, we will not be able to determine the model for this painting.

This month-old discussion, "Who modelled for Frederick Sandys’ painting 'Queen Eleanor'?" has attracted 17 responses. Betty Elzea, the acknowledged expert, says "I have not found any clues as to who she is, but my feeling is that she was a Norwich girl" (3 May). She sees the same model in Queen Eleanor, 1858 (Cardiff), Mary Magdalene, c.1859 (Delaware), Oriana, 1861 (Tate) and King Pelles Daughter, 1861 (Lanigan and Tate).

The National Museum Wales inclines to agree that we will not be able to determine the model (1 June). On this basis, I recommend that we close this discussion unless Andrew Greg, my fellow group leader has further thoughts.

Jacinto Regalado,

It is simply my opinion and others are entitled to differ, but while this young woman might be based on a real person, I still think it could be a pre-Raphaelite tronie.

Thank you Simon. I agree that the discussion, and in particular Betty Elzea's expert opinion, must lead us to conclude that current knowledge makes it unlikely we can identify a named model for Sandys' image of Queen Eleanor.

Howard Jones,


During Queen Victoria's reign a fashion craze began for wearing dresses coloured purple or mauve following William Perkin's discovery of the first coal tar dyes for these colours. He took out his patent for his dyes in 1856 and purple and mauve dresses were being worn by the Royal family at the London Trade Fare in 1862.

As this painting is dated to 18588 could the mauve dress being worn here by Queen Eleanor reflect this new popular fashion for dresses of this colour?

I repeat my recommendation of 5 June that this discussion now be closed. Please see my summary above. We have the agreement of the collection and Andrew as my fellow group leader.