Anyone researching a painting by Roelant Savery will almost inevitably encounter Kurt J. Müllenmeister’s critical catalogue raisonné of 1988 and its 1991 supplement. Yet recent auctions show why simply finding a number is not enough. Some entries contain reservations, others have subsequently been corrected, and paintings outside the catalogue continue to appear on the market. That is precisely where the catalogue raisonné starts to become interesting.

A Savery listed as no. 102, yet still only ‘attributed to’

On 15 September 2026, a relatively modest painting will come under the hammer at Artcurial in Paris: Scène de chasse dans un sous-bois de montagne, oil on oak panel, measuring 38 by 50 centimetres. The estimate is €3,000–4,000. At first glance, the literature reference appears reassuring. After all, the painting is listed in Kurt J. Müllenmeister’s catalogue raisonné of Roelant Savery as no. 102.

Yet Artcurial does not offer the work unreservedly as a Savery. The cataloguing reads “Attribué à Roelandt Savery”. More interesting still is the note accompanying the Müllenmeister number: according to the auction house, the provenance and dimensions given in the catalogue raisonné are incorrect, while Müllenmeister himself expressed reservations about the attribution, particularly because of the handling of the animals.[1]

This small painting is therefore almost a textbook example of how a catalogue raisonné should be read in practice. Anyone who merely checks whether a work has a number would miss precisely the most important information here. The entry actually documents an art-historical position subject to reservations.

The standard reference behind the market

For the painted oeuvre, the central reference work remains Kurt J. Müllenmeister’s Roelant Savery. Kortrijk 1576–1639 Utrecht. Hofmaler Kaiser Rudolf II. in Prag. Die Gemälde mit kritischem Oeuvrekatalog, published in 1988. The substantial volume runs to 427 pages and is explicitly devoted to the paintings.[2] In 1991 it was followed by the 87-page Neues und Ergänzungen zum Oeuvreverzeichnis, Freren 1988.[3]

The fact that a book is now more than thirty-five years old does not mean that the market has abandoned it. Quite the opposite. Auction houses still refer explicitly to Müllenmeister for both major and less important works by Savery.

Nor has recent scholarship replaced it with a new printed standard catalogue of the paintings. The body of work currently attributed to Savery is estimated at approximately 300 paintings and 250 drawings, while works from lesser-known collections continue to emerge on the art market.[4] That last point is crucial. Müllenmeister is a standard reference, but not an endpoint.

A number is not an authentication code

An even sharper example than the forthcoming Artcurial lot is a case handled by Sotheby’s in January 2021. Landscape with swans near a waterfall, a signed panel measuring 31.4 by 39.7 centimetres, carried an estimate of $30,000–50,000, at the time approximately €25,000–41,000. The literature reference read: Müllenmeister 1988, page 263, no. 143A.

But then came the decisive qualification. Sotheby’s stated that Müllenmeister had included the painting with reservations because he had insufficient photographs and information at his disposal.[5]

It is precisely this kind of wording that shows why a catalogue-raisonné entry cannot be reduced to “it is listed” or “it is not listed”. The quality of the original documentation matters. Did the compiler physically examine the painting? Were only old black-and-white photographs available? Was the provenance known? Were the dimensions and signature checked?

With another Savery, we can even see how such research changed over time.

In November 2018, Lempertz sold Landscape with a Tower and Bridge with Waterbirds in the Foreground, signed and dated 1625, for €322,400 including buyer’s premium against an estimate of €180,000–200,000. The large canvas, measuring 68.5 by 99.5 centimetres, is listed by Müllenmeister as no. 162.[6] Lempertz reported that Müllenmeister had assessed the painting for his 1988 catalogue on the basis of photographs.

From the main volume to the supplement

The numbering itself also deserves attention. The market uses both straightforward numbers and numbers with an A suffix. Anyone who automatically reads these codes as a single simple chronological sequence will soon run into difficulties. A good example is Landscape with ruins and a variety of birds, a signed panel probably dated 1614. Sotheby’s referred not to the main 1988 volume, but explicitly to Müllenmeister’s 1991 supplement.

More than thirty years after its publication, the supplement is still cited verbatim as a literature reference. And Sotheby’s did more than simply reproduce no. 235A. The auction house noted that Müllenmeister had linked a comparable composition in the 1988 catalogue to a work in the Kunsthistorisches Museum in Vienna, whereas according to Sotheby’s it is in fact a different painting.[7]

Here we see the catalogue raisonné functioning as a critical reference work should: not as an untouchable final text, but as a point of departure that can be corrected by subsequent research.

What the catalogue can tell us about scarcity

The greatest mistake would be to use Müllenmeister only for questions of authenticity. The catalogue becomes far more interesting when we try to place a painting within a particular part of Savery’s oeuvre. Take the flower still lifes.

On 8 June 2021, Dorotheum in Vienna sold Mixed flowers in a glass vase with a lizard nearby, a signed painting on copper measuring just 31 by 24.5 centimetres. Müllenmeister published it as no. 292, dated it to around 1612 and reproduced it in his book. The estimate was €200,000–300,000; the realised purchase price was €247,300 including buyer’s premium.[8]

Dorotheum described the work as one of only 25 known authentic flower still lifes by Savery. This places the copper within a relatively small and highly recognisable part of Savery’s production.[8]

The difference between €13,000 and €322,000

In March 2025, Artcurial sold Taureaux, perroquet et autres animaux dans un paysage rocheux, a small oak panel measuring 23.5 by 26.5 centimetres, for €13,120 against an estimate of €10,000–15,000. The work was listed in Müllenmeister as no. 116, had according to an old label been exhibited in Ghent in 1954, and had a Belgian provenance extending back to the De Heuvel collection.[9]

It therefore possessed precisely the attributes that make a superficial checklist look attractive: a catalogue-raisonné number, an old provenance and exhibition history. Yet its market value was in an entirely different league from that of the flower still life or the large bird paradise sold by Lempertz in 2018 for €322,400.

That difference is analytically more revealing than the highest auction price in itself. A Müllenmeister number does not flatten the hierarchy of the oeuvre. Size, pictorial type, ambition, condition, date, composition and quality continue to matter.

A recent Sotheby’s cataloguing makes that distinction even more concrete. In December 2025, the auction house offered a small circular panel entitled Geese preening on a bank by a marsh, frogs in the foreground, swans beyond with an estimate of £20,000–30,000. According to Sotheby’s, Müllenmeister no. 141 dates the work to around 1604 and regards it as the pendant to another circular panel depicting a cow, which sold at Koller in March 2021 for CHF 128,400.[10]

And what if a painting is not in Müllenmeister at all?

That is where things become interesting as well.

In November 2022, Lempertz sold a Forest Landscape with Deer and a Cow, oil on panel, measuring 42.5 by 47.3 centimetres. The auction house catalogued it as Roelant Savery, estimated it at €10,000–15,000 and sold it for €11,970 including buyer’s premium. Yet the literature section explicitly stated that the work is not included in Müllenmeister’s catalogue raisonné.[11]

This does not mean that an omitted work is thereby declared “authentic”. But the example does show why absence from a printed catalogue is not the same thing as an independent judgement on authenticity. A work may have been unknown at the time of publication. It may fall into a category not covered by the catalogue. New information may have emerged. Or later specialists may reach a different opinion from the author of an earlier catalogue.

A book published in 1988 can remain authoritative without having closed off the future.

How auction houses actually use Müllenmeister

When we place recent auction catalogues side by side, what stands out above all is the variety of ways in which the same standard reference can function.

Sometimes Müllenmeister appears as a brief literature reference: author, page, number. Sometimes the entry determines the dating. Sometimes it provides an earlier or original title. Sometimes it reveals that two paintings were regarded as pendants. And in the most interesting cases, the cataloguing records that Müllenmeister expressed reservations, had only seen photographs, or relied on information that later proved incorrect.

What strikes us most about Savery, therefore, is not simply how often auction houses refer to the same publication. More important is how seriously they sometimes engage with the information behind the number. Careful cataloguing does not simply read: “Müllenmeister no. 102.”

It explains that no. 102 contains an incorrect provenance and that the author had doubts about the animals. It states that no. 143A was included conditionally because of inadequate photographic material. And with no. 235A, it corrects an identification involving a painting in a Viennese museum. These are precisely the details that transform a catalogue raisonné from a list of numbers into a research tool.

What should a collector actually check?

With a Savery, serious examination therefore does not begin with the question of whether the word “Müllenmeister” appears somewhere in an auction catalogue. The first step is to establish which publication is being cited: the main volume of 1988 or the supplement of 1991. The exact number must then be matched against the title, medium, dimensions, signature and any illustration. The next question is whether the entry contains comments or reservations and what documentation it was based on. Earlier provenance should be compared with later sources; the same applies to dimensions and museum identifications.

In short, the number is a starting point, not a conclusion.

The painting behind the number

Yet the recent market also shows why it is dangerous to confuse the authority of a catalogue raisonné with a simple certification system. No. 102 can accompany an attribution subject to reservations. No. 143A can be based on insufficient photographs. No. 162 may have been physically examined by the author only after publication. And a painting without a number can nevertheless appear at auction as a Savery.

More than three decades after publication, Müllenmeister’s work remains the common reference language for the paintings of Roelant Savery. That alone makes the book indispensable for anyone working with his oeuvre.

Do you have a particular interest in this artist or in this subject?

Do you follow this artist, or have you come across a work you would like to know more about? We regularly research works appearing on the market and are always happy to exchange views on quality, provenance, art-historical context and previous auction results. You can contact us directly via info@antonello.art.

Notes and Sources

[1] Artcurial, Maîtres anciens & du XIXe siècle, Paris, 15 September 2026, lot 151, “Attribué à Roelandt Savery, Scène de chasse dans un sous-bois de montagne”; literature: Kurt J. Müllenmeister, 1988, p. 243, no. 102.

[2] Kurt J. Müllenmeister, Roelant Savery. Kortrijk 1576–1639 Utrecht. Hofmaler Kaiser Rudolf II. in Prag. Die Gemälde mit kritischem Oeuvrekatalog, Freren: Luca Verlag, 1988.

[3] Kurt J. Müllenmeister, Neues und Ergänzungen zum Oeuvreverzeichnis, Freren 1988, 1991.

[4] Olga Kotková, “Roelant Savery’s Wondrous World”, The Burlington Magazine, vol. 166, no. 1454, May 2024. The review states that approximately 300 paintings and 250 drawings are currently attributed to Savery and notes that works from lesser-known collections continue to emerge.

[5] Sotheby’s, Master Paintings Part II, 30 January 2021, lot 519, Roelandt Savery, Landscape with swans near a waterfall, estimate $30,000–50,000; Müllenmeister 1988, p. 263, no. 143A, described by Sotheby’s as having been included with reservations because of insufficient photographs and information.

[6] Lempertz, sale 1118, Cologne, 17 November 2018, lot 1533, Roelant Savery, Landscape with a Tower and Bridge with Waterbirds in the Foreground, estimate €180,000–200,000; realised price €322,400 incl. buyer’s premium; Müllenmeister 1988, p. 271, no. 162.

[7] Sotheby’s, Old Masters Day Sale, London, 8 December 2022, lot 423, and Old Master Paintings Day Auction, London, 7 December 2023, lot 28, Roelandt Savery, Landscape with ruins and a variety of birds; Müllenmeister 1991, pp. 63–64, no. 235A. Estimates £80,000–120,000 and £40,000–60,000 respectively.

[8] Dorotheum, Old Master Paintings I, Vienna, 8 June 2021, lot 26, Roelant Savery, Mixed flowers in a glass vase with a lizard nearby, estimate €200,000–300,000; purchase price €247,300 incl. buyer’s premium; Müllenmeister 1988, p. 341, no. 292.

[9] Artcurial, Maîtres anciens & du XIXe siècle, Paris, 27 March 2025, lot 219, Roelandt Savery, Taureaux, perroquet et autres animaux dans un paysage rocheux, estimate €10,000–15,000; sold for €13,120; Müllenmeister 1988, p. 251, no. 116.

[10] Sotheby’s, Old Master Paintings Day Auction, London, 4 December 2025, lot 120, Roelandt Savery, Geese preening on a bank by a marsh, frogs in the foreground, swans beyond, estimate £20,000–30,000; Müllenmeister 1988, no. 141.

[11] Lempertz, sale 1209, Cologne, 19 November 2022, lot 1517, Roelant Savery, Forest Landscape with Deer and a Cow, estimate €10,000–15,000; realised price €11,970 incl. buyer’s premium.