Layers of an Oil Painting: From Support to Varnish

What are the layers of an oil painting?

An oil painting can include a support, size, ground, preparatory drawing or paint, image-forming paint layers, and varnish. These describe different functions, not a compulsory number of coats. Materials and sequences vary between works and within one picture. Use a painting's published technical record to establish its actual structure; do not clean, sample, unframe, or illuminate an artwork to investigate it yourself. Examination and treatment belong to qualified conservators.

This guide concerns the structure described in museum catalogues, not a studio recipe. A diagram is useful when it helps you ask what lies beneath a surface. It becomes misleading when it tells you what must be there before anyone has examined the object.

How does the basic sequence fit together?

Read the following table from the material carrying the painting toward the viewing surface. Some rows may be absent, overlap, or contain several applications.

These are categories of function, not six numbered strata guaranteed to appear in a microscopic sample. “Ground” may name a preparation built from multiple coats. A drawn contour occupies a line rather than an entire rectangle. Paint applied to a face need not continue beneath the adjacent sky.

Start with this distinction: a painting has an overall construction, but the sequence at a particular point is local.

What is the difference between the support and size?

The support carries the work. Canvas is a fabric support; a panel is a rigid one. Neither word identifies all the preparations above it. A label saying “oil on canvas” therefore leaves many structural questions unanswered.

Size is a material applied in preparation, not a measurement of the painting. In discussing canvas paintings, conservation scientist Stephen Hackney explains the protective role of glue size between canvas and oil-containing layers. His account also shows why this is not an uncomplicated barrier: the glue and the other materials have their own ageing and mechanical behaviour. See Understanding Structure, Changing Practice.

For a catalogue reader, the practical distinction is between the carrier and an applied preparation. If a report identifies canvas, glue size, and an oil-containing ground, do not compress them into “paint directly on fabric.” Equally, the historical description is not an instruction to prepare glue or treat an existing canvas.

Our museum-label guide explains how to use the accession number to find the longer record behind an abbreviated medium line.

What do ground, gesso, and imprimatura mean?

A ground prepares the support for painting. Its colour can contribute to the appearance of the paint above it. The National Gallery's ground definition notes both coloured preparations and double grounds. A double ground means two ground layers; it does not mean a double-sided painting.

Gesso identifies a more specific material in historical accounts. The Gallery describes gypsum with an animal-glue binder, while noting that northern European panel preparations generally used chalk or another calcium carbonate instead of gypsum. Its gesso entry also distinguishes a coarser preparation, gesso grosso, from a finer one, gesso sottile.

Thus “white ground” alone is not a chemical identification. It tells you less than a report that actually identifies the mineral and binding material.

An imprimatura is a thin, translucent application of colour over the ground. The Gallery's definition describes its role in reducing absorbency and sometimes providing a middle tone. It is preparation for the image, not necessarily the first depiction of a figure or landscape.

When reading “priming,” follow the author's description of location and composition. Do not add another layer to your notes merely because a second term appears. Ask whether the report describes a separate application or is explaining an already named preparation.

Where do underdrawing and underpainting belong?

Underdrawing establishes a composition through preliminary outlining on the prepared support. That is the scope of the National Gallery's underdrawing definition.

Underpainting refers instead to early painted development beneath later applications. The distinction is chiefly between graphic planning and painted development, not between two obligatory manufacturing stages. A report may describe preliminary contours in one passage and broad painted masses in another.

Nor does the term “oil painting” settle the binder of every preparatory layer. The National Gallery's survey of Northern European painting, 1400–1550 discusses linseed and walnut oils alongside egg tempera, used predominantly for underpainting in the material it surveys. That is evidence for mixed practices within a defined historical corpus, not a formula for every northern painting.

Our separate guide to what underdrawing is covers its detection and interpretation. Here, the important point is its position within a larger construction: discovering a drawing tells you about planning, but not automatically about every paint application above it.

Are all the coloured layers glazes?

No. The binder is what holds pigments together to form paint; in oil paint, that binding material is a drying oil. The National Gallery distinguishes the binder from the coloured pigments. Layer structure and paint composition are related questions, but they are not interchangeable.

A glaze is a transparent or translucent coloured paint film that modifies what lies beneath without simply hiding it. The Gallery's glaze entry explains that its thickness can affect tonal variation.

A scumble is a thin application lighter than the colour beneath; the Gallery explains that its translucency comes from thin application rather than the inherent optical character of its constituents. See its scumble definition.

Neither term should be assigned merely because a reproduced painting looks luminous. The question is how a particular passage was constructed. A painting can combine more covering paint with translucent applications rather than consist entirely of successive glazes.

This also separates a glaze from varnish. The former is coloured paint participating in the image; the latter is a surface coating with a different function. “Transparent layer” is insufficient to identify either one.

What does an examined painting actually show?

Raphael's Ansidei Madonna supplies a specific alternative to a generic diagram. The National Gallery reports a wooden panel, a gesso ground, and oil-based priming. Examination identified walnut oil as the paint binder. Saint John the Baptist's red robe combines several semi-opaque applications with translucent glazes. Those are findings about this painting, not deductions from its appearance. See Studying Raphael: pigments and medium.

The same research describes green landscape paint beneath grey architectural paint, establishing a change during execution. A paint cross-section, a microscopic view through a mounted sample, supplied evidence of that sequence.

Notice what the example does not give us: one layer count for the entire altarpiece, or a universal “Raphael method” to apply to unrelated works. The robe and the altered background answer different local questions.

The useful habit is to attach each technical finding to its location. Write “green landscape beneath grey architecture,” rather than “the painting has a green underpainting throughout.”

Did different traditions follow different orders?

Yes, but broad regional labels need qualification. The historical distinction between gypsum and chalk preparations is one meaningful difference. The documented coexistence of oil and tempera in northern practice is another. Neither produces a complete recipe for an entire region.

Venetian painting also resists a simple opposition between drawing and colour. In their study of Titian's technique to about 1540, Jill Dunkerton and Marika Spring explain that infrared reflectography overturned the belief that he invariably began directly in paint without drawing. Infrared reflectography is imaging that can reveal some underlying features using infrared radiation.

The authors also report revisions, or pentimenti, evident in technical images and overlapping colours in cross-sections. Drawing did not prevent subsequent alteration.

This is an examination-based correction to a general account of artistic practice. It is more useful than replacing one slogan with another, such as declaring that every Venetian painter followed the same preliminary design procedure.

How do written instructions differ from physical evidence?

A treatise is a written account of methods. It can document a recommended practice, terminology, or material choice. A technical examination addresses what survives in a particular object. Those are different forms of evidence, even when they agree.

For reading purposes, separate three statements:

• A historical writer recommended a preparation.
• Researchers identified that preparation in an examined painting.
• A modern author proposes that the painting followed the written instructions.

The third statement is an interpretation connecting the first two. It needs an argument, not just similar vocabulary.

The examples above rely on published museum examination findings and definitions; they do not reconstruct a recipe from a historical treatise. When a catalogue invokes one, record the named text and the basis for applying it to that object. Leave uncertainty visible.

Is varnish always the final original layer?

Varnish is a coating that affects surface reflection and can protect the paint. Historical varnishes include several kinds of material, and they can discolour with age, as the National Gallery's varnish glossary explains.

A coating described in a present-day condition report is not automatically the artist's own application. Look for the report's distinction between original materials and later interventions. The varnish line in a diagram records a possible function, not an assurance of an untouched surface.

Do not test a coating with water, solvents, rubbing, or scraping. Do not remove a frame or use ultraviolet or infrared equipment to investigate an artwork. A qualified paintings conservator must assess examination and any treatment; historical material descriptions are not safe-use instructions.

How can you make a useful layer note?

Use this original reading exercise with a published museum record, not with physical tests on an artwork:

1. Copy the support and medium exactly, including uncertainty.
2. List only preparations the technical text actually identifies.
3. Attach each paint-layer observation to its stated location.
4. Record the evidence: document, surface observation, imaging, or sample.
5. Keep the author's conclusion separate from your own question.

For example, a fictional note reading “white layer beneath blue” does not establish gypsum, a particular white pigment, an artist, or a date. The next task is to find what the report says about that layer, not to supply a familiar material name.

The working-methods collection develops that approach: define the term, locate the evidence, and retain the limits. A good layer diagram should make a painting easier to understand without making it appear simpler than the evidence allows.

Sources

• National Gallery: Ground, Gesso, and Imprimatura.
• National Gallery: Underdrawing and Binder.
• National Gallery: Glaze, Scumble, and Varnish.
• National Gallery: Studying Raphael: pigments and medium.
• National Gallery: Methods and Materials of Northern European Painting, 1400–1550.
• National Gallery: Titian's Painting Technique to c.1540.
• Getty Conservation Institute: Understanding Structure, Changing Practice.