Imagining Artist Methods
RuyschBouquet of Flowers · 1708All 16

A studio guide · Amsterdam, 1708

Bouquet of Flowers, pressed and painted

Imagine you are Rachel Ruysch, in your forties, the most sought-after flower painter in Amsterdam, in the year the Elector Palatine names you his court painter. This guide follows how you might have built up one bouquet, from a dark ground to the last petal. Then it shows the one step that is documented: a real Red Admiral's wings, pressed into wet white paint and lifted off, leaving their scales behind.

No full technical study of this canvas has been published. The butterfly print was described by a conservator at the Doerner Institut in Munich in 2024; the ground follows another Ruysch painting, analysed at the Mauritshuis. The other stages are inferred, and each chapter opens with a note on what its evidence is. The slider steps through seven stages; stages 1 to 6 are reconstructions, made partly with AI.

The wing pressed: today

Stage 7 of 7

A real Red Admiral's wings pressed into the white and lifted off, the scales left in the paint, the tips glazed. The painting as it is, 92.3 by 70.2 cm.

Rachel Ruysch, Bouquet of Flowers (Blumenstrauß), 1708. Oil on canvas, 92.3 × 70.2 cm. Bayerische Staatsgemäldesammlungen, inv. 430, Staatsgalerie im Neuen Schloss Bayreuth. AI-generated Stages 2 to 6 are built from this painting: stages 2 and 3 trace and mask it, stage 4 combines it with a layer generated with Prodia, and stage 5 removes the insects with Prodia. Base image: a public-domain scan, Wikimedia Commons.

Your training

EvidenceThe 2024 Munich exhibition booklet and its biographies; the Met's record of the 1692 portrait. Your birth date and much of your early life rest on later biographies.

AI-generated reconstruction: a painter's room in an Amsterdam canal house with a tall leaded window, a woman in a dark dress and white cap seen from behind at an easel with a dark flower painting, a glass vase of flowers on the table.
AI-generatedA painter's room in Amsterdam around 1700, as it might have looked. Where you worked is not recorded.

You are Rachel Ruysch, born in The Hague in 1664. Your father, Frederik Ruysch, is an anatomist and botanist; your mother, Maria Post, comes from a family of painters and architects. In 1666 the family moves to Amsterdam.

At about fifteen your parents pay for an apprenticeship with Willem van Aelst, the best-known still-life painter in the city. You stay about three years, until his death in 1683. From him you take a way of building a bouquet: not symmetrical, but swept on a diagonal from lower left to upper right, with flowers turning in and out of the light.

In 1693 you marry the portrait painter Juriaen Pool; you will have ten children. In 1701 the two of you join Pictura, the painters' confraternity in The Hague. You are its first woman member. Women cannot join the painters' guild in Amsterdam.

Your father's cabinet

EvidenceThe exhibition booklet on Frederik Ruysch and on Otto Marseus van Schrieck. That you used specimens from your father's collection in this painting is not documented.

AI-generated reconstruction: a natural-history cabinet with shelves of glass jars of preserved plants and small reptiles, drawers of pinned butterflies under glass, shells, coral and a microscope by a window.
AI-generatedA collector's cabinet in Amsterdam around 1700, as it might have looked.

Your father lectures to the surgeons' guild, trains the city's midwives and, from 1685, directs Amsterdam's botanical garden. He is famous across Europe for his preparations: bodies and organs injected with tallow, wax and cinnabar so that they look alive, kept in jars of spirits. He opens his collection to visitors as a museum; in 1717 he sells it to Peter the Great.

Flowers, insects and reptiles were part of the household. Flower pieces of this period often gather in one bouquet plants that do not bloom at the same time: a convention of the genre, not a record of one vase.

From Otto Marseus van Schrieck, who had worked with Van Aelst in Italy and died in Amsterdam in 1678, you take more than motifs. He pressed real butterfly wings into his paintings, and so, sometimes, will you. The Munich exhibition suggests you may even have used specimens your father bought from his estate; it says no more than that.

Court painter, 1708

EvidenceThe exhibition booklet on your appointment and on Zacharias Conrad von Uffenbach's visit of 1711; the Bavarian State Painting Collections' record, which says this canvas came from the Düsseldorf gallery in 1806. No document read names it as painted for the Elector.

By now your paintings are ordered up to a year ahead. When the traveller Zacharias Conrad von Uffenbach visits in 1711, he finds two finished pieces only because they are waiting for their buyer, Pieter de la Court in Leiden, who has paid 1,500 guilders for the pair. Your brushwork is so fine, the Munich exhibition reports, that by your own account you finish only three or four paintings a year.

In 1708 Johann Wilhelm, Elector Palatine, makes you his court painter, with a yearly salary and gifts in exchange for one painting a year. You do not have to live at his court in Düsseldorf; you stay in Amsterdam. Later he and his wife, Anna Maria Luisa de' Medici, stand godparents to your youngest son and come to Amsterdam for the baptism.

A dark ground
Like many flower painters of your generation you set the bouquet against near-black; you will lighten your backgrounds only after 1735.
A diagonal
Tulips and a poppy at the top right, the guelder rose low at the left, the wheat falling over the ledge: Van Aelst's sweep, made fuller.
A ledge
A brown stone ledge with an orange, a snail and a grasshopper, so the eye has somewhere to stop.
One real insect
A Red Admiral by the wheat, at the viewer's eye level, made from a real butterfly's wings.

Was this the Elector's painting?

Probably. It is dated 1708, the year of your appointment, and it reached Munich in 1806 from the Düsseldorf gallery, where Johann Wilhelm's collection hung. But the record does not say that he commissioned it, or that it was the first of your yearly paintings for him.

What is known of the materials

EvidenceMarlena Schneider (Doerner Institut, 2024) for the white of the print and the vermilion and black on the wing tips; Albrecht and Meloni (2025) for the ground of a different painting, the Mauritshuis Vase with Flowers of 1700. Nothing else about this canvas's materials has been published.

The canvas is 92.3 by 70.2 centimetres. Its ground, its paint layers and its pigments have not been published. Conservators and scientists in Munich and elsewhere studied your technique for the 2024–25 exhibition, but what has appeared so far is a short article on the butterflies and the titles of talks.

The closest published evidence is your Vase with Flowers of 1700 in the Mauritshuis, eight years earlier and also on canvas. Its ground is two dark brown layers of earth pigments, chalk and lead white, the upper with more and finer lead white. The cross-section here assumes the same for this canvas and says so.

The Munich exhibition also showed films on X-ray pigment analysis of your paintings and on an early use of Prussian blue, a pigment first made in Berlin in the first decade of the 1700s. Which painting had it was not stated; nothing here assumes it.

What has been found, and where. The swatches approximate each colour in oil.

Lead white

Painted roughly in the shape of the butterfly and left wet, so that the wing's scales stick to it.

Where
Under the Red Admiral
Found
Described for the method, 2024

Vermilion

Glazed over the scales of the wing tips, with black.

Where
The Red Admiral's wing tips
Found
Described in 2024; method not stated

Black

A deep black, also glazed over the scales of the wing tips. Which black is not stated.

Where
The Red Admiral's wing tips
Found
Described in 2024; method not stated

Butterfly scales

Not a pigment: the coloured scales of a real wing, left in the paint when it was lifted off.

Where
The Red Admiral, the whole wing
Found
Microscope, 2024

Earths, chalk and lead white

The two dark brown ground layers of your 1700 painting in the Mauritshuis. Assumed here, not found here.

Where
Ground of the 1700 Vase with Flowers
Found
Cross-sections, published 2025

Ground, lay-in and background

EvidenceAlbrecht and colleagues (2023) on the Mauritshuis flower pieces, of which only the summary was read; the Munich booklet on the c. 1682 Flower Piece. Nothing on this canvas: every step in this chapter is inferred.

AI reconstruction: A bare canvas covered evenly in dark brown.
GroundDark brown. AI-generated
AI reconstruction: Thin pale lines on the dark ground marking the ledge and the outlines of the main flowers.
Lay-inThin pale lines. AI-generated
AI reconstruction: The dark background and ledge painted, the bouquet's place still bare ground with pale lines.
BackgroundBouquet left in reserve. AI-generated
AI reconstruction: The large flowers finished, the leaves and small flowers as flat olive patches.
Large bloomsThe rest blocked in. AI-generated
Stages 1 to 3 are drawn from the finished painting: its masks, its traced contours and its own background. Stage 4 adds a layer generated with Prodia. They show plausible stages, not scans of real layers.
  1. Prime the canvas dark

    A dark brown ground, as under your 1700 bouquet. It gives the dark background a head start.

  2. Place the bouquet

    Mark the ledge and where the main blooms go. A study of nineteen flower pieces in the Mauritshuis found that Jan Davidsz. de Heem's circle began with abstract patches of colour, circles and ovals, but that Van Aelst and you planned your compositions differently. How exactly is not in the part of the study read for this guide, so the lines shown are a placeholder.

  3. Paint the background around it

    Brush in the near-black background and the stone ledge, leaving the place of the bouquet bare. That is one common sequence; whether you worked this way on this canvas is not known.

  4. Change your mind if you must

    In your Flower Piece of about 1682, an X-ray fluorescence scan made for the exhibition shows a stone arch drawn in vermilion in the background, then abandoned and painted over. Nothing like it has been reported for this canvas.

Your earliest painting in Munich, the Flower Piece of about 1682, was conserved for the exhibition. Its mercury map, mercury being the metal in vermilion, showed the outline of the arch you gave up. The same painting has butterfly prints of which only fragments survive; the restorers decided not to fill the losses, so that the prints stay legible.

Paint the bouquet

EvidenceThe painting itself and the exhibition booklet. The order of the three steps is inferred; the plants are identified by eye, not from a botanical study.

The finished canvas shows no seam between passages: the Munich exhibition calls your brushwork invisible. So the order below is a reasonable guess, not a finding.

1 2 3

AI-generatedAfter step 3: insects with the brush

  1. 1The large blooms
  2. 2Leaves, small flowers and wheat
  3. 3Insects with the brush

The large blooms

AI reconstruction after step 1: Tulips, roses, the poppy and the guelder rose finished; leaves, small flowers, wheat and the orange only blocked in. Inferred.
After step 1 AI-generated
Close-up of a white tulip flamed with deep red stripes, its petals curling open against the dark.
The striped tulip at the top.

Begin with the flowers that carry the arrangement: the striped tulips, the roses, the red poppy at the top and the white guelder-rose ball at the left. Their placing sets the diagonal sweep from lower left to upper right that you learned from Willem van Aelst.

Nothing published says in what order you painted the flowers of this canvas. Large blooms first is a common sequence and the one shown here.

Close-up of a full white rose with pale shadows in its folded petals, a blue morning glory above.
The white rose at the centre.
Close-up of a large deep red rose, heavy and half-open, at the right of the bouquet.
The red rose at the right.

Leaves, small flowers and wheat

AI reconstruction after step 2: Every plant finished, and the orange; no insects yet. Inferred.
After step 2 AI-generated
Close-up of an orange on the stone ledge among dark green leaves, with small white flowers at the left.
The orange on the ledge.

Then fill in around them: leaves turning in and out of the light, the small flowers, the stems, and the ears of wheat that hang down over the ledge. An orange sits on the stone at the left.

Your own estimate, as the Munich exhibition reports it, was three or four paintings a year. How long this one took is not recorded.

Close-up of ears of wheat hanging over the edge of the ledge, with a small butterfly beside them.
The wheat over the ledge.
Close-up of a small snail with a pale spiral shell on the brown stone ledge, leaves above it.
A snail at the left end of the ledge.

Insects with the brush

AI reconstruction after step 3: The other insects painted with the brush, and the outline of the Red Admiral filled with wet lead white. The method is documented; when you did it is inferred.
After step 3 AI-generated
Close-up of a large pale green grasshopper with spread, veined wings on the ledge.
The grasshopper at the lower right.

Last come the animals: a grasshopper on the ledge, the snail, moths and butterflies among the flowers. Whether any of these, besides the Red Admiral, is a print rather than painted has not been reported.

For the Red Admiral, paint its rough outline in lead white and leave it wet. Chapter VII follows what happens next.

Close-up of a brown and cream butterfly with spotted wing edges under a striped tulip.
A butterfly under the tulip: painted or printed is not reported.

The butterfly print

EvidenceMarlena Schneider, Doerner Institut (2024), who examined the Red Admiral under the microscope; the Munich booklet on the method. This is the best-documented step in the guide.

AI-generated reconstruction: a close-up of a hand with fine tweezers pressing a red admiral butterfly's wing onto a patch of wet white paint on a dark, unfinished flower painting, beside a painted ear of wheat.
AI-generatedImagined: a wing pressed into wet lead white. Whether you used tweezers, or your fingers, is not recorded.

Paint the rough shape of the butterfly in lead white and leave it wet. Take a real wing, from a dead butterfly, and press it into the paint; then lift it away. The coloured scales stay stuck in the white, and with them the pattern of the wing, down to the fine veins and hairs.

Not every scale will take, and the body, head and antennae cannot be printed. Paint those in with the brush, and fill gaps in the pattern. Varnish and oil also make some scales go clear, which is another reason for touching in.

Here you chose a Red Admiral, a native butterfly, and set it on a stalk of wheat at the bottom of the picture, at the viewer's eye level. Under the microscope the whole wing proves to be covered in scales. You can see the strongly patterned underside of one wing, with its typical half-ring markings, and part of the upper side of another: you printed two wings one over the other. You seem to have added only a little paint to the pattern. Over the scales of the wing tips you glazed a deep black and a bright vermilion.

Reconstruction: a patch of white paint the shape of a butterfly on the ledge beside the wheat.AI-generatedStage 6: the outline in wet lead white.
The Red Admiral in the painting today, with dark wings, orange-red bands and white spots.Today: the printed wings.

The print was found in 2024, during the research for the exhibition, and the Munich conservators call it nearly perfectly preserved. That is rare. Butterfly scales are fragile, and most prints in old paintings survive only in fragments, worn by light and by rough cleaning. In your Flower Piece of about 1682 the restorers took off the yellowed varnish under a stereomicroscope so as not to lose the remaining scales.

Is every butterfly here a print?

Not as far as anyone has reported. The published find is the Red Admiral. The other butterflies and moths in the bouquet may be painted, printed or both; they have not been described.

Afterwards: Düsseldorf, Munich, Bayreuth

EvidenceThe Bavarian State Painting Collections' record; the Munich booklet; Pigage's catalogue of the Düsseldorf gallery (1778). No condition report or treatment history for this canvas has been published.

The painting entered the Bavarian collections in 1806 from the gallery in Düsseldorf, Johann Wilhelm's picture gallery, one of the finest in Europe. Some of the paintings you made for him appear in the drawn wall views of the Düsseldorf and Mannheim galleries; whether this one does has not been checked for this guide. Today it belongs to the Bavarian State Painting Collections and hangs in the Staatsgalerie in the Neues Schloss at Bayreuth.

For the exhibition Rachel Ruysch: Nature into Art, shown at the Alte Pinakothek in Munich, then at the Toledo Museum of Art and the Museum of Fine Arts, Boston, in 2024–25, conservators looked closely at your Munich paintings, and the butterfly print on this one was found. A fuller account of their technical findings has not yet been published.

The image this guide is built on is a scan of a printed reproduction from 2007, before that examination. Its colours are warmer and its darks flatter than the painting's.

What is still not known

This canvas's ground, and whether there is any drawing under the paint. The order in which you painted it, and how long it took. Its pigments, beyond the vermilion and black on the butterfly. Whether the other butterflies are prints. Whether it is signed, and where. Whether the Elector commissioned it. What the X-ray and pigment studies made for the 2024–25 exhibition found.

How Ruysch thought

  • Sweep the bouquet. A diagonal from lower left to upper right, learned from Van Aelst and made fuller.
  • Set it in the dark. A near-black background, so the petals come forward.
  • Give the eye a ledge. A stone edge with fruit, a snail and a grasshopper where the bouquet ends.
  • Let nature print itself. Real scales from a real wing, pressed into wet white paint.
  • Then paint what will not print. Body, head and antennae by brush, and glazes over the scales.
  • Work slowly. Three or four paintings a year, by your own account.

Sources