By Aron Lewin
Edgar Degas’ Portrait of a woman was, and is, a phenomenon.
The oil-painting of the woman was acquired by the Felton Bequest for £2275, and given to the National Gallery of Victoria in 1937.
It is, according to NGV senior curator Ted Gott, a “freely painted image of a plumpish woman in a black dress and bonnet. The unidentified sitter leans forward, poised on the brink of conversation, her eyes alive with interest.”
“The concentration with which she listens to an off-stage interlocutor has been captured with deft, quickly applied brushstrokes that accord with the artist’s painting style of the late 1870s.”
The discolouration on the face of the woman was the cause of some consternation, and it was met with mixed reviews when it came to Australia.
William Fielding Wannan wrote for The Argus that it was “deplorable that the National Gallery continues to pay enormous sums for pictures that are either of dubious authorship or inferior examples of a master’s work.”
The director of the National Gallery of Victoria (NGV) at the time, JS MacDonald, also said the “discolouration on the face of the subject is [only] due to photography … it is not an especially good Degas, but they are almost impossible to come by.”
However, MacDonald did not know what the discolouration entailed.
In 1952, a violet ray device was used, which highlighted surface cracks in the painting. These cracks were scanned using an infrared camera.
The images revealed a second figure behind the portrait, but art conservators did not have access to imaging techniques which would allow them to clearly see who that hidden person was.
From The Tasmanian Advocate, March 8, 1952:Degas portrait was a "double". “Infrared and violet ray photography has disclosed that “Portrait of a Woman,” painted by the celebrated French artist Degas, who died in 1917, aged 83, was painted over the top of another portrait, also of a woman.Gallery officials said yesterday that the “Portrait of a Woman," which was valued at £4000, had lost no value by the discovery. Mr. Darly Lindsay, director of the National Gallery, said it was possible that Degas has had no other canvas available on which to paint the portrait.
This technology is now available.
The painting was taken to the Australian Synchrotron – a football field sized partial accelerator – where it underwent rapid, non-invasive and high definition x-ray fluorescence (XRF) elemental mapping.
The Synchrotron generates a source of incredibly intense light – more than a million times brighter than the sun – and directs it into an x-ray beam that is 1/10 the diameter of a single human hair.
Every point of the painting was scanned through the x-ray beam for less than a millisecond, and an XRF detector, situated just 2mm from the painting surface, recorded the elemental content. The 31.6 megapixel images resulting from the scan show, layer by layer, the metal elements contained in the paint.
After 33 hours in the X-ray Fluorescence Microscopy (XFM) beamline, all the information was loaded onto a computer screen and a colour reconstruction of the hidden figure was revealed, based on the analysis of the paint pigments that had been gained.
For example, the Synchrotron researchers wrote that “Fe [iron] and Mn [manganese] are co-located in the hidden sitter’s hair, strongly suggesting the use of the brown pigment umber.”
As predicted by newspapers across Australia in 1952, a hidden female figure was discovered underneath the discolouration.
She is believed to be French model Emma Dobigny whom Degas painted more than once.
The synchrotron's revelation of the hidden Degas painting caught the attention of art-lovers around the world, with intense media interest from The New York Times to The Bangkok Post.
Australian Synchrotron scientist Dr Daryl Howard, who led the work, said the combination of Degas star-power and the attractiveness of the hidden figure meant the story made a huge splash.
“Because Degas is so well known internationally the name that goes with the artwork does all the work for you. A lot of [the attention] comes with who the artist is,” he said.
“A story I like to recount is an episode of Antiques Roadshow, where one of the hosts said ‘if you have a painting of a pretty woman, it goes a long way in [enhancing] its value'. So I think that’s what really caught the public imagination.”
A local mystery: McCubbin’s The North wind
While the Degas discovery swept through the press, the mystery behind Frederick McCubbin’s The North wind received less fanfare.
However, head of conservation at the NGV Michael Varcoe-Cocks said the story was more complex and interesting, but harder to tell on the front page of a newspaper
Under a bright cerulean blue sky, the painting shows an exhausted family trawling through the dusty road and barren surroundings. The man is shielding his face and the family seems to struggle with the heat and wind. The horse, head down, looks shattered.
The painting is known to have undergone a number of alterations. According to an eBook written by Phip Murray and Michael Varcoe-Cocks, it is suspected Frederick McCubbin’s son Louis made changes to The North wind some years after the artist's death.
“The type and style of interventions made to it align with those made to [his] other paintings … It is well documented that Louis posthumously altered a number of McCubbin's work. These alterations included reformatting canvases, restoring areas of damage and adding a new signature or date,” Varcoe-Cocks and Murray write.
“McCubbin’s original composition appears to have been cut down on the left side and also extended at the bottom; the reason for the trimming is unknown, but may have been due to damage … Regrettably, this section is now permanently lost.”
The use of an unknown cleaning agent had removed layers of paint, and areas of it were overpainted in order to cover up the damage. Throughout the years, The North wind had also become discoloured with an accumulation of grime and dirt.
Believing it to be an unfair rendering, Varcoe-Cocks took the painting off display. He intended to restore it quickly, but the process ended up taking about 10 years.
“Along with technical research, we do a lot of historical research,” he said.
“For the first couple of years I spent a lot of time in archives, searching around for letters by the artist and trying to identify … where it was first exhibited.
“We also wanted to know who owned it before the NGV owned it, because the [Felton Bequest] purchased the work in 1941 but our records don’t show who we purchased it from.
“So, there’s about 50 years before it came to us that we can’t account for."
Michael Varcoe-Cocks and Dr Daryl Howard talk in detail about the work done on The North wind.
The fragile and enormous painting – about one metre by a metre and a half – was taken to the synchrotron for analysis. After scanning through the layers of paint, the scientists mapped the distribution of elements and found zinc had been used to restore different parts of the painting.
“We can see, for example, where there was a tear, they’ve filled it with zinc white and then retouched the area, very broadly. It’s an interesting way of mapping out, across the surface, the original and non-original layers,” Varcoe-Cocks told ABC's Catalyst last year.
Pigments containing zinc were also found on the signature, indicating that it was added later by somebody else.
After two days at the synchrotron, the high-resolution image of the elemental map that was produced showed an entirely different scene to the one on the surface.
McCubbin’s initial version of The North wind had a lush green landscape. There was a different rendition of the horse underneath, and a woman and baby had replaced a solitary man sitting on the cart.
According to the NGV eBook, “the painted-out male figure was reimagined on the right side of the work”.
The Synchrotron website states that the changes were made “amid the centenary of settlement commemorations in 1888 and the emergent idea of the ‘Aussie battler’.”
“I was fairly sure that the signature and date were applied later on, which raised all sorts of questions … as to whether the painting was an original McCubbin painting or a fraudulent work,” he said.
"So we had to go through the work of confirming that the signature and date weren’t original, and also that the whole painting was in fact done by Frederick McCubbin.
“The material data acquired through the work we did [at the NGV] and also the final imaging done at the synchrotron confirmed all of these questions for us.”
Dr Howard said the XFM beamline mapped the pigment distribution on The North wind quite nicely.
“The scene had changed so drastically from its original look. For instance, the family is now in a dusty, outback scene with the horse looking tired with his head down. In the underlying painting, the horse is looking alert with his head high,” he said.
“There must have been quite a change in McCubbin's way of thinking, or maybe the times had changed from when he originally painted it. For me, that shows a bit of insight into his character.
“It was an impressive bit of detective work to pull it all together.”
The test case: Arthur Streeton’s self portrait
A mystery of a different sort was hiding behind an Arthur Streeton canvas.
The impressionist artist had covered up a self-portrait with white lead paint in order to reuse the canvas.
It was not used at a later date, and retained by his family. Streeton’s grandson, Oliver, lent the canvas to researchers from the synchrotron, NGV and CSIRO, who chose to work on it as a test case in 2010.
Research program leader Deborah Lau wrote in The Conversation that the experiment allowed the team to perform in-depth analysis of the painting without damaging the artwork.
“Before performing any work on the Streeton painting, we applied historic paints actually used by Streeton to canvas swatches and irradiated them with a dose 1000-times greater than the actual scanner would expose the painting to,” she wrote.
“At no point was damage observed. This is important because if art galleries are going to give us access to paintings, they want to know we aren’t going to damage them, and rightly so.”
After scanning the painting, Dr Howard and the team of scientists at the Synchrotron mapped the elemental distribution within the painting.
“What’s interesting is that the self portrait is hidden underneath a layer of lead white paint, which is a really strong absorber of x-rays,” Dr Howard said.
“With normal techniques by gallery heads, they could tell there was a portrait underneath that was quite faint and not of very good quality. It was a good test of the technique to see if we could get the hidden image out and reveal it.
“It turned out successfully, and was one of very first publications we did on a painting. It led to the McCubbin and Degas studies, and really paved the way for where we are at today.”
Dr Howard says that he can understand why Streeton covered the portrait up.
“I am no art expert, but this probably isn’t his best work. He did submit a self portrait to, I believe, the Art Gallery of New South Wales … so it might have been a test work before he settled on the final self portrait.”
Since the Streeton experiment in 2010, Dr Howard said the XFM beamline detector had become more advanced.
“To collect the data for every pixel, you’d have to stay at the point for a second or a half second or so. Imagine if you wanted to do a billion pixels, it would take a billion seconds. That’s a lot of time,” Dr Howard said.
“Now, instead of having to stop and dwell for that amount of time … we are constantly moving the painting through the x-ray beam, and any point the painting is under the x-ray beam for less than a millisecond, So roughly 1000 times faster than we could in early days.
“This has opened up large areas for mapping, and it’s a lot safer because … you always want to put the least amount of x-ray beam on the painting. With this detector we can do that amazingly well.”
First-class Marksman, by Sidney Nolan (1946). Picture: Art Gallery of NSW.
Sidney Nolan’s Ripolin paint cans
Rather than identifying a hidden feature in a painting, this conservation project was about Sidney Nolan’s choice of paint.
The iconic Australian artist was one of the first artists to experiment with commercial house paints and other then-unusual media such as PVA and spray cans.
A study by Dr Paula Dredge showed that Nolan used the semi-synthetic based nitrocellulose paint in 1941 and synthetic-based alkyd paint in 1942.
She writes that following his conscription to the war in 1942, he worked with an alkyd lacquer Dulux, but there was little available and production halted until after the war.
Fortunately, Nolan found a large quantity of an imported paint – the Ripolin brand of brightly coloured glossy enamel house paint.
However, it dried slowly, forcing Nolan to leave his paintings in front of a heater, slowing down his work.
“Ripolin is like quicksilver I can [see] us cooking it over a fire or leaving it out under the rosemary all night to see what secrets can be found in it,” Nolan reportedly said in 1943.
But he was attracted to the paint, in part because Pablo Picasso famously used it from about 1912. Nolan later said: “Why should I bugger around and torment myself and a wise man says its healthy paint. OK, so Ripolin it had to be.”
Nolan used the paint almost exclusively from 1943 until his departure from Australia to the UK in 1953.
Paintings that he made using the Ripolin brand of enamel paint include Wimmera landscape, a 1943 self portrait, Bathers, First class marksman and The Camp.
Dr Dredge – who was completing her PhD on the contents inside Nolan’s studio – received more than 100 items and 35 cans of Ripolin from Sidney Nolan’s daughter Jinx in 2006.
“No one really knew what Ripolin paint was … everybody knew that Nolan used it and Picasso used it, but there was a lot of confusion in the literature talking about what the constituent materials in it were,” she said.
“This is important because we like to catalogue our material. We will always say ‘oil on canvas’ or ‘oil on board,’ but there’s been confusion about how to catalogue Ripolin."
The contents from 18 different colours of Ripolin paint cans were tested at the Art Gallery of New South Wales. The paint was found to be oil-based, rather than synthetic, and also contained a hard tree resin called copal.
She took samples of the white paint to the synchrotron for infrared scanning after discovering it contained zinc oxide.
“We see in 20th century paintings there was a shift … [where painters] used zinc oxide in preference to lead for their base white paints.
“The reason why [zinc oxide] was in Ripolin was that there was an increasing awareness that lead-based house paints were poisonous.
“However, it was problematic in terms of art conservation. There’s a reaction between zinc oxide and the fatty acids within the oil [which makes the paint] incredibly brittle and weak as it ages. They form a crack pattern, which is obviously a concern.
“So one of the nice things about the infrared mapping we were able to do at the synchrotron was to try and map that reaction, and try to find out what the impetus for that reaction was.”
The scans tracked a transformation from zinc oxide to zinc stearate (so-called zinc "soap", which repels water). This knowledge, further developed by Queensland Art Gallery & Gallery of Modern Art head of conservation Dr Gillian Osmond, opened up some work that was "really pivotal in the conservation field", Dr Dredge said.
She said that her work had given researchers a better handle on the experimental materials Nolan used.
“He came from a background as a commercial artist where he was spray painting signage and making shop displays from the age of 13,” she said.
“The paints he chose were those that he knew and used as a commercial artist … now we think using house paint is no big deal. But prior to and through the Second World War this was unusual.
“He was one of the earliest to use the first synthetic paint called alkyd. He used PVA [polyvinyl acetate glue] in 1957 when only a handful of artists did that and took up painting with spray cans before street art and graffiti became really big.
“I don’t think its been fully appreciated how avant-garde Nolan was with the materials he used.”
A question of time: Charles Conder's While daylight lingers
Details relating to the origin of Charles Conder’s While daylight lingers are not known, but there might be clues in its lining.
Conder, an English-born painter, designer and lithographic apprentice for the Sydney Illustrated News, produced several major landscape, portrait and silk-based ornamental paintings.
In 1890 the prominent Heidelberg School figure produced While daylight lingers. The impressionist piece shows a silhouetted shepherd herding sheep in a paddock. Trees in the background offer little shade under a hazy and cloudy sky.
Dr Howard said the painting was lined with newsprint, and he had hoped the XFM beamline would reveal what was written on it and allow them to date the painting accurately.
“Unfortunately we were not able to image that,” Dr Howard said.
“This is probably because most of the elements in the newsprint were carbon, which we can’t detect at all. I was hoping there’d be iron in the newsprint, which we would be able to identity.
“But we were still able to successfully scan the painting and get the elemental composition of it. This really helps the conservators understand what pigments the artist was using, and what changes were made to the painting as well.
“We are hoping to write the study up in the next year or so, but these things take a lot of time.”
Into the future …
The synchrotron and the NGV are currently working on several other paintings, but Dr Howard says that they have to keep quiet until the galleries are ready to announce.
“I don’t want to jump the gun too far ahead, but we have some exciting things to come,” he said.
“It’s going to be really great.”