I. Factors to Consider when Choosing a Varnish

From Wiki

Back to Paintings Chapter List

The information presented on the Paintings Conservation Wiki is the opinion of the contributors and does not imply endorsement or approval, or recommendation of any treatments, methods, or techniques described.

This entry is in the process of being corrected

Painting Conservation Catalog

Author: Wendy H. Samet
Date: Submitted August, 1996
Compiler: Wendy Samet


A varnish can serve several functions, technical as well as purely visual. The decision to apply varnish to a picture is made after careful consideration of many factors and cannot be reduced to a formulaic approach. If a varnish is to be applied, many further decisions such as type, method of application, and desired final appearance must be considered. The following is intended as an aid to the decision-making process.


The first question to take into account when considering varnishing a painting is whether or not varnish is appropriate for a particular work of art.

1. Most traditional Western paintings prior to the late 19th century were meant to have a surface coating of some kind.
2. Since the middle of the 19th century, some artists from the Impressionists on have felt that varnish was unacceptable for certain of their works.
a) Paysage à Chaponval by Pissarro, formerly in the Jeu de Pomme, has the following inscription by Pissarro on the verso: “Veullies ne pas vernir ce tableau, C. Pissarro. “This painting remains unvarnished (Stoner 1985, 87–8).
b) If it is known that an artist felt strongly that one work should not be varnished, one cannot extrapolate that no works by that artist are to be varnished.
c) If the Impressionists had known about thin, stable modern synthetic varnishes and minimal application methods, would they have felt as strongly opposed to varnishing?
d) There may be legitimate reasons for a conservator to choose to varnish a painting which the artist had intended to be unvarnished including:
(1) visual changes in the appearance of the painting due to aging;
(2) visual changes in the painting due to removal of a previously applied varnish; and
(3) the need to apply a varnish as a barrier. (See Barrier or Protective Considerations, below.)
3. Varnish may be unacceptable for paintings with certain intentional physical characteristics (Research usually indicates that these paintings were originally unvarnished).
a) Intentionally Matte or Underbound Paint (e.g., some Cubist paintings, some German Expressionist paintings, Tuchlein paintings, etc.)
Note: John Richardson observed in the New York Review of Books that many Cubist pictures had been ruined by varnishing, among other conservation processes (Richardson 1983).
b) Paintings with Intentional Matte/Gloss Differences (e.g., some paintings by Juan Gris)
c) Paintings with Intentionally Exposed, Absorbent Grounds or Supports (e.g., some paintings by Leger, Vuillard, or Cezanne)
d) Paintings with Certain Collage Elements (e.g., newspaper)


1. Historical Considerations

Customary Practices of the Time, of the School, of the Artist

a) Most traditional Western paintings prior to the late 19th century were intended to receive some final finishing layer.
b) Early Western paintings in egg tempera may or may not have been varnished.
“The picture was born of successive layers, of which the last two are varnishes applied by the painter himself. The first of these consists of egg white diluted in water…” (Del Serra 1985, 4) “… between the tempera and the final covering of vernice grassa, a thin covering of egg white was added as soon as the painting was finished. The point was not only to maintain the brilliance of the egg tempera, but also to avoid distortion of the color and uneven chromatic change. Without this intervening layer of egg white, some colours by their very nature would have absorbed more of the vernice grassa coating than others” (Del Serra 1985, 4).
Cennini mentions the use of glair as an alternative to vernice liquida if time was short (Bomford et al. 1989, 183).
“At the end of his section on panel painting Cennino gives relatively brief instructions on how and when to varnish a panel with ‘vernice liquida’… recipes in a slightly later Bolognese Manuscript show that basically it consisted of semi-hard resins such as sandarac dissolved and boiled in linseed oil.” (Bomford et al., 182–3.)
There is documentation that the altarpiece of San Pier Maggiore in Florence was sent to Sant Maria Nuova in 1371 to be varnished. Remnants of an apparently original varnish of this type were found (Dunkerton et al. 1990, 63) and there is no evidence of an intervening glair varnish (Dunkerton et al. 1990, 65).
This varnish would have been highly glossy (Bomford et al. 1989, 183).
Descriptions of varnishing paintings in these writings [handbooks written by craftsmen] are thus good evidence that paintings were indeed varnished, though not always, and not necessarily by, or in the workshop of, the person who painted them” (section author's italics) (Dunkerton et al. 1990, 67).
We can be quite certain that early Italian panel-paintings were intended to look dazzlingly bright. Rich, pure pigments in egg tempera, most notably vermilion and ultramarine, were combined with the solid metallic appearance of pristine gilded backgrounds, and were engraved, stippled, and tooled for added luxury. … Were varnishes part of this medieval aesthetic? The limited evidence that survives suggests that, in general, they were not. It seems fairly clear that the gold leaf backgrounds were never intended to be varnished. Cennino's description of how tooled and stippled gold was meant to sparkle confirms this. Often the original desire for deep-coloured burnished gold and lighter reflective stippling has been reversed by varnishing. However, whether the painted surfaces were to be varnished remains less certain. Examples of never-varnished early Italian panels have been discovered. But a few also survive with early glair or oleo-resinous varnish layers. Cennino mentions the use of oil varnishes. Nevertheless, an overall consideration of medieval aesthetic tastes would cause us to question whether these varnishes were ever intended to be used on a regular basis” (Hoeniger 1994, 12).
c) Early Western paintings in oil or oil-and-tempera technique were probably varnished with boiled oil or fossil resins which impart highly saturating and glossy coatings. (See Section II., Traditional Artists'Varnishes and Section B.4., below.)
d) Most traditional Western oil paintings from the 16th century on have been intentionally varnished.
(1) In general, for traditional Western paintings, paintings that employ glazes, or those with rich deep tones, varnish with a high degree of saturation is required.
(2) Spirit varnishes, “soft” resins dissolved in solvent, came into use sometime in the 18th century.
(3) Mastic seems to have been the most commonly used spirit varnish until the second or third quarter of the 19th century. (See Section IV.B.2., Mastic.)
(4) Damar varnish was introduced in the second or third quarter of the 19th century. (See Section IV.B.2., Damar; for a more complete discussion of historic varnishes, see Section II, Traditional Artists'Varnishes.)
(5) “Many assume that dammar is a good example of an original varnish. In fact, dammar was apparently introduced only in the nineteenth century. Although it may have been popular in Germany, it was not widely adopted in Britain, where mastic was the only spirit varnish recommended for paintings throughout the nineteenth and well into the twentieth century. As well, spirit varnishes have not been used exclusively; oil-resin varnishes have had an even longer history of use. Although oil-resin varnishes are certainly not a desirable alternative for a paintings conservator today, our view of what could have constituted an original varnish must include these materials as well.
If we are going to adopt natural resins on the basis of an ‘authentic’ appearance and search for alternative synthetic materials which better replicate the appearance of natural resins, then we should consider more closely what was actually in use at a given time and place.
Are mastic and dammar actually as interchangeable in appearance as has been assumed by some twentieth-century researchers, and how do these spirit varnishes compare with oil-based varnishes?
Before we are able to explore the subtleties of perceptual differences in colour, gloss, and saturation of the most popular traditional varnishes, including those made using oil, we need to provide ourselves with examples” (Carlyle 1994, 9).
e) Some paintings are meant to be left unvarnished. Among others, Impressionist paintings and 20th-century paintings will each have their own set of requirements regarding whether or not to varnish, degrees of saturation and gloss, and surface uniformity or a lack thereof, that will need to be taken into account on an individual basis. (See discussion in Sections b)(1)–(4), above.)

2. Artifactual Considerations

a) Preservation of an Original Surface Coating
When presented with a painting with its original varnish—especially if that varnish is of historic importance, intact, and not disfiguring—the conservator may choose to preserve it.
(a) Reforming a varnish by spraying or brushing it with solvents can impart improved saturation and a more even gloss.
(b) An even appearance may be difficult to achieve.
(c) The solvents required may affect the painting.
(d) Reforming an original varnish changes its structure.
(a) for visual reasons
i) to resaturate
ii) to achieve an even gloss
iii) drawbacks listed under “Reforming” above may apply to varnishing over an older coating
(b) for protective/preservation reasons
i) An unstabilized natural resin varnish exposed to UV light will continue to degrade. Applying an acrylic varnish with UV absorbers over such a varnish may help to stabilize it (section author's italics) (Bourdeau 1990; de la Rie 1992, 71).
ii) Recoating an original varnish may help protect it from atmospheric dirt, abrasions, or possibly even some forms of vandalism.
iii) See drawbacks listed above.
(c) solubility concerns
i) The conservator may want to revarnish using a varnish that is similar to the original.
ii) The conservator may want to apply a varnish that can be removed from the original varnish in the future.
b) Original, Intentionally Temporary Varnishes (a rare occurrence)
Sometimes an original varnish was meant to be a temporary varnish, such as an egg-white coating used over a fresh oil paint applied before the paint had time to dry properly (Stringari 1990; Peres 1990).
(a) This varnish may not have the visual properties considered appropriate for a final varnish.
(b) This varnish may be visually disturbing.
i) It may be discolored.
ii) It may prevent proper saturation by a later varnish.
(a) for visual reasons
i) to resaturate
ii) to achieve an even gloss
(b) for protective/preservation reasons
Recoating an original varnish may help protect it from atmospheric dirt, abrasions, or possibly even some forms of vandalism.
(c) solubility concerns
It should be possible to remove the new varnish from the original.
c) Preservation of an Original Unvarnished Surface
When a conservator treats a painting with an intentionally unvarnished surface, important decisions must be made.
(1) An unvarnished surface is vulnerable to:
(a) atmospheric dirt becoming embedded
(b) abrasions
(c) vandalism
(2) Given the range of stable synthetic varnishes and the variety of methods of application, varnishing might be an option.
(3) Other options may include:
(a) controlling the environment
(b) limiting handling, movement, and travel
(c) glazing and/or placing the painting in a vitrine

3. Aesthetic Considerations

The Artist's Intent in a Picture

Acceptance or mitigation of alterations by time and/or past intervention(s)

Note: Aesthetic considerations are complex, yet crucial to the appropriate selection of varnish materials and their means of application.

a) Sources for Selecting Aesthetically Appropriate Varnishes
(1) Curators or art historians may provide valuable information regarding the probable intended visual appearance of a picture.
(2) Artists' statements and/or contemporary accounts about the intended appearance of a work of art.
(a) Where an artist's intents or preferences are stated, vagaries of language and its change over time and cultures may render such words less than clear.
(b) Time and past intervention may render an artist's intent unachievable (e.g., a porous painting meant to remain unvarnished cannot be perfectly recovered if varnished in the past).
(3) Photographs can yield helpful clues to a painting's previous coating history and appearance, but photographs may be easily misinterpreted due to lighting conditions at the time of the photograph, aging of the photograph, over- or underexposure of the photograph, use of early, black-and-white, nonpanchromatic films, retouching of photographs, or printing of the photograph.
(4) Technical analysis, especially of an oeuvre or school.
(5) Careful visual reading of the painting itself will likely give clues as to its aesthetic requirements.
(a) Construction
i) possibly do not coat
porous paint
exposed porous ground
collage elements
ii) possibly use highly saturating varnish
to detail in areas of dark or black colors
when glazes have been used
(b) Current condition and appearance of the painting including possibly:
i) pigment changes
ii) increased transparency
iii) damages
iv) problems from prior cleanings
(6) Personal preference of the owner, curator, or dealer

4. Barrier or Protective Considerations

All resins discussed in this volume are permeable to light, oxygen, and atmospheric moisture. The following are some considerations.

a) To isolate original layers from later retouchings or from other coatings.
b) To mitigate the deteriorative effects of ultraviolet light and/or oxidation by adding layers with UV absorbers (Bourdeau 1990; de la Rie 1992, 71). See Section VII.A., Phenolic Antioxidants, Stabilizers, and UV Absorbers.
c) To isolate the paint layer from:
Perhaps most famous is the case of Picasso's Guernica, which was protected from vandalism in the form of red spray paint by a moderately heavy and continuous brushed coat of Acryloid® B-72 it had received several years before. The spray paint and varnish were immediately removed with xylene (Aviram, telephone conversation 1996).


1. Considerations for Selection

a) Composition
The composition of the varnish, any additives or stabilizers, and delivery solvents must be known.
b) Reversibility
A varnish “must remain highly soluble so that it may be removed from the work of art at any time in the future” (Feller, Stolow, and Jones 1985, 153).
c) Stability to Color Change
“The varnish should be transparent, colorless and resistant to discoloration” (Feller, Stolow, and Jones 1985, 153).
d) Adhesion to Substrate
“Any varnish with a tendency to cleave or shear from a particular paint surface or other varnish layer cannot provide an adequate visual appearance.” (Feller, Stolow, and Jones 1985, 153).
e) Protection
The varnish should protect the painting from atmospheric dirt and moderate abrasion.
Note: According to René de la Rie, varnishes are much more susceptible to chemical reaction and degradation in their liquid state and should therefore be kept in dry resin form until just before use. This is particularly critical if stabilizers are to be included in the varnish (de la Rie and McGlinchey 1989, 144). (See Section VII.A., Phenolic Antioxidants, Stabilizers, and UV Absorbers.) This clearly has implications for commercially prepared varnishes which are purchased already dissolved in usually undisclosed solvent mixtures and which may claim to contain various stabilizers.

2. Visual Properties

In general practice, the properties of gloss and saturation work in concert with one another, although they are separate phenomena. Gloss is a surface phenomenon that describes the degree to which light is scattered or not as it reflects off the surface of a varnish or paint film. Saturation refers to the degree of depth of color a varnish can render. Although exceptions can be made, those varnishes which are the most saturating tend to be among the most glossy.

Note that matte or unvarnished surfaces are one extreme and that highly glossy and saturating surfaces are another. Many subtleties of gloss and saturation are appropriate and achievable within these extremes. It is up to the conservator to determine what is appropriate and to master the techniques necessary to produce the desired results.

a) Gloss
“Gloss can be described as the degree to which a surface approaches a mirror-like surface” (Eastaugh 1984, 10). Most conservators probably think of it as the degree of shininess that a varnish imparts to the surface of a painting. Conversely, matteness can be defined as having little gloss. Gloss is entirely a surface phenomenon. Judging the inherent glossiness of a particular varnish and/or comparing it to the glossiness of another is a difficult prospect. Gloss can be measured in several ways. (For an excellent article on the subject, see Eastaugh 1984, 10–14; see also Feller, Stolow, and Jones 1985, 138–42). Any comparisons must, of course, be made not only on the same type of measuring device but on the same piece of equipment, under identical conditions, with samples identically prepared by the same person. In real life, as will be noted throughout this volume, there are many ways of modifying and applying each varnish to affect gloss, including diluent, dilution, brushing/spraying/rubbing/etc., sequencing, adding matting agents, and manipulation after drying. Furthermore, the gloss of a varnish is likely to change with time.
In determining gloss, it is important to understand the two forms of surface reflection used to describe the various types of gloss. “Specular reflection is light observed from a mirror-like surface where the angle of reflection is exactly equal (but opposite) to the angle of the incident beam of light. … Diffuse reflection is where the incident beam of light is evenly scattered in every direction from the reflecting surface” (section author's italics) (Eastaugh 1984, 10).
Terms used in characterizing certain aspects of gloss are defined below:

Sheen is similar to gloss but is defined as the shininess at grazing angles. In practice, it is used for the classification of surfaces (e.g., paper) having a low level of gloss that can be seen only at low angles.
Luster is contrast gloss, the ratio between the amounts of light specularly reflected.
Distinctness-of-image gloss is, as its name suggests, an indication of how clearly or sharply specularly reflected images can be seen, indicating slight variations of the mirror image around the angle of reflection.
Absence-of-bloom gloss is a combination of these last two forms of gloss, and is a measure of haze or milkiness adjacent to reflected high-lights. This forms a description of the amount of diffuse reflection adjacent to the specular image.
Surface uniformity gloss is independent of reflectance, being instead a qualitative assessment of the freedom from visible non-uniformities such as texture. As it is a graded assessment of the surface quality, it cannot be measured, just as the Munsell chart is intended as a subjective assessment of colour” (Eastaugh 1984, 10).

b) Saturation
The specific saturation of a varnish is dependent upon an interaction of three different characteristics: the varnish's ability to level; its ability to wet onto a specific surface, be it paint or a lower layer of varnish; and its refractive index.
“… if a varnish dries with a rough surface, its contrast gloss and distinctness of image gloss will be markedly affected … if a solvent-type varnish forms an immobile gel at a point when considerable solvent still remains, it will tend to form a surface which will follow the irregularities of the paint underneath the varnish” (Feller, Stolow, and Jones 1985, 140–1). If a varnish does not level, it will diminish distinctness-of-image gloss and scatter light. The ability to level well is directly related to the size of the varnish molecule or polymer. Polymeric varnishes such as the acrylics are quite large in size as compared to the natural resins, ketone resins, and low molecular weight varnishes such as Arkon® P-90 and Regalrez® 1094. During drying (solvent evaporation) these high polymer varnishes form an immobile gel with relatively high concentrations of solvent left in the drying varnish such that they tend to conform to the surface of the underlying layer rather than leveling. Note the relatively high viscosities of polymeric varnishes at say 20% concentration in solvent as compared to that of the lower molecular weight varnishes. (See charts on Molecular Weight and Viscosity Grade below.) In contrast, the lower molecular weight varnishes continue to flow and therefore level, even when the concentration of remaining solvent is quite low. The effect of the relative lack of leveling of the acrylics can be somewhat mitigated by using slower evaporating solvents. See Section X.A. “Varnishing Acrylic Emulsion Paintings,” for specific recipes developed for this purpose.
“The loose term ‘wetting’ is usually used to describe the phenomena that take place at the interface of a solid and a liquid” (de la Rie 1987a, 9). In this case the solid is presumed to be the paint layer and the liquid the applied varnish. “Lack of contact between a varnish and a paint surface would result in light scattering due to air pockets. No scientific studies on the degree of contact between the picture varnishes and surfaces of paintings are known to the author” (de la Rie 1987a, 8). Empirical experience demonstrates that penetration into porous paint is characteristic of varnishes prepared with polymers of low viscosity grade. As in leveling, molecule size plays a big role in the ability of a varnish to wet onto a surface. Viscous, large molecule varnishes cannot insinuate themselves into the tiniest interstices of a weathered paint film both due to their size and due to the fact that they form an immobile gel before achieving optimum contact with the paint film. Wetting of a paint surface is also related to the problem of chemical compatibility.
This is the ability of a material to bend light entering it from another material. In practical terms, the higher the refractive index of the resin, or the closer it is to that of aged linseed oil (approximately 1.57 according to Feller), the less the light incident upon the painting will be scattered; thus more light will get through the varnish, into the paint film (into the medium and pigment particles), back through the varnish, and into our eyes (Kushel, unpublished notes to students).

3. Physical Properties

a) Molecular weight and Molecular Weight Distribution (Polydispersity)
In general, there is a direct correspondence between increased molecular weight and increased viscosity of a resin in solution at any given percent. Therefore, a 10% solution of Paraloid® B-72 with a MWn (number average molecular weight) of 11,397 and MWw (weight average molecular weight) of 65,128 can be predicted to be considerably more viscous than a 10% solution of damar, MWn 488 and MWw 1,361. Experimentation has shown that the viscosity grade (the viscosity in centipoise of a solution of the resin in toluene at 21°C and at a concentration of 20% by weight) of Paraloid® B-72 is 29 while that of damar is 1.3 (de la Rie 1987a, 5). For a discussion on the relationship between molecular weight and leveling, see 2.b)(1) above.
The ratio of the weight average molecular weight and the number average molecular weight (MWw/MWn) for a given varnish indicates the molecular weight distribution or polydispersity for that varnish. Natural resins tend to have high polydispersities due to the presence of small polymeric fractions (de la Rie 1987a, 5). This wider range of polymer sizes within the material leads to a greater ease in brushability.
b) Viscosity
The literature generally gives the intrinsic viscosity for varnishes. Intrinsic viscosity is measured by plotting viscosities of solutions at various concentrations and projecting the plot to 0–1% varnish. The measurement of viscosity is dependent on both the solvent chosen and the temperature at which the tests are conducted (Maines, Verbal communication 1995). See also 3.a)(2), above, for a discussion of the role of molecular weight in leveling.
c) Refractive Index
A varnish with a high refractive index will tend to give better saturation than one with a lower RI, but RI is only one component of saturation and is generally thought to be less important that either leveling or wetting ability (Feller, Stolow, and Jones 1985, 143).
d) Glass Transition Temperature (Tg)
A varnish with a low Tg near room temperature will tend to retain airborne dirt. Varnishes with high Tgs tend to be more brittle and scratch more easily than those with a lower Tg.
e) Brittleness, Flexibility, and Hardness as Predictors of Scratch Resistance/Durability and Toughness
These characteristics are related and can be generally predicted from several of the properties found in the materials section on each individual varnish. A varnish with a high Tg is more brittle and therefore more easily scratched than one with a low Tg. A large molecule or high-polymer varnish is more flexible and therefore less likely to scratch and/or shear from a paint surface. Hardness tests, while they cannot be directly compared unless done by the same person under identical conditions, may predict the scratch resistance of a particular varnish.

4. Chemical - Chemical Structure as a Predictor of Stability

“Resins for picture varnishes should … be of low polarity and free of functional groups that render them unstable. … The absence of photochemically active functional groups in the resin molecules, such as ketone groups and carbon-carbon double bonds, insures resistance to autoxidative degradation” (de la Rie 1990, 168). Other functional groups which may contribute to increased autoxidation by free radical production include “ether groups and tertiary carbon atoms. Hydrogens in a-positions to double bonds and those attached to tertiary carbon atoms are more easily abstracted than others” (de la Rie 1992, 69).

5. Comparative Charts

a) Factors Affecting Gloss and Saturation
(The higher the molecular weight, the more viscous the solution, the less well it tends to level)
Number Average Weight Average
Molecular Weight Molecular Weight PolydispersityMn/Mw
Damar 488a 1361a 2.79a
Gum Mastic 460a 1929 4.20a
Paraloid® B-72 11,397b 65,128b 5.72b
Paraloid® B-67 10,960b 44,764b 4.08b
PVAAYAA 1,691b 88,567b 2.80b
PVAAYAC 5848b 15,185b 2.60b
PVAAYAF 51,370b 117,697 2.29b
MS2A® 620a 1144a 1.85a
Laropal® K80 531a 734a 1.38a
Arkon®P-90 620a 934a 1.51a
Regalrez® 1094 630 900
Escorez® 391/349a 652/518a 1.67/1.48a
a de la Rie and McGlinchey 1990a, 168.
b de la Rie 1987a, 4.
“The viscosity in centipoises of a solution of the resin in toluene at 70°F (21°C) and at a concentration of resin of 20% by weight” (Feller, Stolow, and Jones 1985, 126–7).
Damar 1.3
Gum Mastic 1.8
Paraloid® B-72 29.0
Paraloid® B-67 18.0
Laropal® K80 1.2
Arkon® P-90
Regalrez® 1094
a Data taken from Feller, Stolow, and Jones 1985, 122.
The closer the refractive index is to that of oil paint, i.e., the higher it is, the less light is scattered at the varnish paint interface and the better the saturation of the paint.
Damar 1.539a
Gum Mastic 1.539a
Paraloid® B-72 1.487b
Paraloid® B-67 1.486c
PVAAYAA 1.4669c
PVAAYAB 1.4669c
PVAAYAC 1.4669c
PVAAYAF 1.4669c
MS2A® 1.518a
Laropal® K80 1.529a
Arkon® P-90 1.522a
Regalrez® 1094 1.519a
Escorez® 1.546/1,548a
a de la Rie and McGlinchey 1990a, 168.
c Information taken from the entries on individual varnishes in this volume.
(Graph taken from de la Rie 1987a, 8)
A 20% gloss and distinctness-of-image gloss (D/I) of some resins applied over substrates of varying gloss
    20% Gloss Substrate with highest gloss Substrate with intermediate gloss Substrate with lowest gloss
Resin/Solvent 20 gl D/I 20 gl D/I 20 gl D/I
Damara/turpentine 90.2 93.8 90.5 88.9 90.4 84.8
Damara/toluene 90.2 96.4 89.9 92.7 90.2 87.3
Mastica/turpentine 90.1 97.5 88.2 91.9 91.2 89.3
Winton® Picture Varnishb 88.5 97.7 88.8 87.6 89.7 85.3
Laropal® K80a/toluene 90.0 91.4 87.9 86.9 88.9 89.7
AYACa/toluene 69.1 80.1 67.5 81.3 67.7 66.1
Paraloid® B-67a/mineral spirits 76.3 98.4 73.9 86.2 36.3 7.5
Paraloid® B-72a/diethyl benzene 77.0 97.7 73.4 79.2 55.0 11.5
Paraloid® B-72a/toluene 76.9 96.3 73.8 77.8 40.9 9.4
All gloss values are in percent.
Gloss values and roughness heights for substrates are, in order of decreasing gloss:
   60° gloss - 64.0, 38.7, and 8.5
   20° gloss - 18.2, 3.1, and zero
   roughness heights - +2.5, +5.5, and 10.5 μm
   a See comments at Tables 1 and 2.
   b A commercial solution of ketone resin I mineral spirits: Winsor & Newton, Inc., Secaucus, NJ 07094.
(a) 60° Gloss


“60° gloss of polymeric varnishes, natural resin varnishes, synthetic low-molecular-weight (LMW) varnishes, and varnishes containing mixtures of LMW resins and polymers, applied over a rough paint surface. The polymer mixed with Arkon® P-90 and Regalrez® 1094 was Kraton® G1650. Elvacite® 2044 was mixed with the aldehyde resin. Polymer concentrations in the mixtures were 5% and 10%.” (Graph taken from de la Rie 1993.)
Mar Resistance (Graph taken from de la Rie 1993.)
Mar resistance films of polymers, natural resins, synthetic low-molecular-weight (LMW) resins, and synthetic LMW resin/polymer mixtures. The polymer mixed with Arkon® P-90 and Regalrez® 1094 was Kraton® G1650. Elvacite® 2044 was mixed with the aldehyde resin. The polymer concentration in the mixtures was 10%.


The lower the Tg, the softer the varnish, and the more likely it is to retain airborne dirt on its surface.
Damar 39.3
Gum Mastic 34.7
Paraloid® B-72 40a
Paraloid® B-67 55a
MS2A® 54.1
Laropal® K80 50.8
Arkon® P-90 35.6
Regalrez® 1094 43.8
Escorez® 48.6/31.5
a Data taken from the entries on individual varnishes in this chapter.
All other data in this chart taken from de la Rie and McGlinchey 1990a, 168.


1. Solvent Choice

a) Recent research suggests that the solvents in which varnishes are applied may have effects on the long-term physical, chemical, and aging properties of these coatings (Hansen et al. 1991; Lawrence 1990).
b) Although relatively little studied by the conservation field, the effects of solvent choice for a particular resin on the physical properties of a film (e.g., gloss and hardness) has long been examined by the paint and coatings industry. Nevertheless, it appears that the effect of solvent choice on the chemical aspects of a coating, such as long-term stability, has not been significantly evaluated by either conservators or industry. While a solvent-type varnish dries by loss of solvent, solvent choice for a particular resin may have a profound and lasting effect on the characteristics of a surface coating. Varying solvent composition may affect numerous aspects of a varnish film, including application characteristics (e.g., rate of drying, brushability, and sprayability), physical properties (e.g., hardness, strength, and elongation, and adhesion to a substrate), and aesthetic properties (e.g., saturation and even discoloration) (Swaraj 1985, 470–534).
It appears likely that solvent choice may also affect the chemical properties of a given film. Polymers in their initial, solid state are long chains that tend to be retracted in tight coils. A thermodynamically favorable or “good” solvent for the resin is one that tends to swell and loosen the coils in random conformation, allowing maximum solvent-polymer interaction. By using a good solvent for initial varnish formulation, one achieves maximum film strength and adhesion to the substrate. In contrast, a “poor” solvent will cause the polymer chains to remain tightly coiled. In general the conservation field has adopted from industry the practice of using a “good” solvent for a given resin in varnish formulation. However, it appears possible that the state in which the resin exists as a dried film (i.e., tight, retracted coils or loose, swollen coils) may have bearing on how the polymeric materials will chemically respond to light and heat upon aging. This would mean that the solvent choice itself would have an effect on the chemical properties, such as long-term stability, of an aged resin.
“… long polymer chains will often crosslink with one another upon aging, making the film less able to swell in the same solvent. It is suggested that a “poor” solvent which allows the polymer chains to remain somewhat retracted, may partially inhibit interchain interactions, such as crosslinking, upon aging … poorer solvent may also impart less strength and adhesion of the film.” (Lawrence 1990, 67–8).

2. To Effect Matte/Gloss, to Affect Degree of Saturation

Although different varnishes have intrinsic qualities of gloss, saturation, and workability, these qualities may be modified by manipulation of brushed application, spray application, manipulation upon drying, buffing, the addition of matting agents, the solvents selected as diluents, and varnish/diluent ratios, in addition to sequencing of different varnishes. These manipulations are discussed within the sections on individual varnishes as well as in Section IX. General Application Techniques, p. 249.

3. Effectiveness as a Barrier

a) Environmental
(1) Varnish has not been found to be an effective barrier or buffer to changes in relative humidity.
(2) Varnish may serve as a sacrificial layer or barrier to particulate contaminates depending on the particular Tg of the varnish. Varnishes with lower Tgs will tend to trap dirt within them, while varnishes with higher Tgs may collect dirt primarily on their surfaces. Brush coats, because they are continuous, may prove to be more effective barriers than spray coats. Ultraviolet absorbers added to acrylic resins may provide some protection for more ultraviolet-sensitive layers below, either ultraviolet-sensitive varnishes or fugitive pigments. The thickness and evenness of the barrier layer are crucial, according to recent tests. See Section VII.A., Phenolic Antioxidants, Stabilizers, and UV Absorbers.
b) Vandalism/Accidents
(1) A varnish may provide some protection from vandalism or accidents. Depending on the type of varnish and its thickness, a varnish may mitigate damage from:
(a) paint
(b) minor scratches
(c) pencil, marker, ballpoint pen, etc.
(2) Varnish will not protect a painting from attack by concentrated acid.
(3) A continuous brush coat may be more effective as a barrier than a thin, spray coat of varnish.


1. Light Levels

Many varnishes, especially natural varnishes and polycyclohexanones which are left unstabilized, are extremely sensitive to degradation by light. Tests indicate that even with the inclusion of a hindered amine light stabilizer such as Tinuvin® 292, these varnishes will discolor and/or degrade unless ultraviolet radiation is removed from the painting's environment. Studies indicate that low molecular weight varnishes such as Arkon® P-90 and Regalrez® 1094 are stabilized against light degradation by the addition of Tinuvin® 292 even in the presence of ultraviolet light. The acrylic and poly(vinyl) acetate varnishes are unlikely to yellow due to photodegradation. An ultraviolet stabilizer such as Tinuvin® 327, used in an acrylic varnish over an unstabilized natural resin, may retard discoloration of the varnish (See Section VII.A., Phenolic Antoxidants, Stabilizers, and UV Absorbers) (Bourdeau 1990; de la Rie 1991). In general, where possible, light levels should be kept below 15 foot-candles and ultraviolet light should be excluded.

2. Temperature

Temperatures in most museum and normal living settings will not greatly affect a varnish. Higher temperatures may cause varnishes with fairly low Tgs to become slightly tacky and retain airborne dirt.

3. Relative Humidity

High relative humidities may cause natural resin varnishes to exhibit a milkiness known as bloom. Varnishes do not protect paintings from changes in relative humidity.

4. Pollution and Particulate Contaminants

A varnish layer may protect a paint surface from airborne particulates by becoming a sacrificial intermediate layer which can be removed and replaced.


Although varnishes themselves are generally not extremely harmful, they are often delivered in solvents which may be dangerous. Material Safety Data Sheets are available for all varnishes and for all solvents. These should be consulted prior to using any varnish to assess personal safety considerations as well as issues of proper disposal.


Aviram, A. 1996. Telephone Conversation (June 11).
Bomford, D., J. Dunkerton, D. Gordon, D., and A. Roy. 1989. Art in the making, Italian painting before 1400. London: National Gallery Publications.
Bourdeau, J. A. 1990. A further examination of the barrier properties of Tinuvin 327 ultraviolet absorber in the protection of dammar films. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the Contributions to the IIC Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 165–7.
Bourdeau, J. A. 1990. 1995. Varnishes: Authenticity and permanence, A Report on the Ottawa Colloquium. In AIC Paintings Specialty Group Postprints, St. Paul Minn, June 9–10, 1995. Washington, D.C.: American Institute for Conservation of Historic and Artistic Works: 10–19.
Carlyle, L.A. 1994. Reproducing traditional varnishes: Problems in representing authentic surfaces of oil paintings. In Varnishes: Authenticity and Permanence, L. Carlyle and J. Bourdeau, eds., Abstracts for the Colloquium, September 19–20, 1994, Ottawa, Canadian Conservation Institute: 9.
de la Rie, E.R. 1987a. The Influence of varnishes on the appearance of paintings. Studies in conservation 32(1):1–13.
de la Rie, E.R. 1987b. Research on picture varnishes: Status of the project at the Metropolitan Museum of Art. In Preprints of the 8th Triennial Meeting of the ICOM Committee for Conservation. Paris: International Council of Museums: 791–6.
de la Rie, E.R. 1992. Stability and function of coatings used in conservation. In Polymers in conservation: Proceedings of an international conference organized by Manchester Polytechnic and Manchester Museum, 17–19 July 1991. N.S. Allen, M. Edge, and C.V. Horie, eds. Cambridge: Royal Society of Chemistry: 62–81.
de la Rie, E.R. 1993. Polymer additives for synthetic low-molecular weight varnishes. In Preprints ICOM Committee for Conservation 10th Triennial Meeting, Washington, DC, USA, 22–27 August 1993. Preprints, Vol. 2. Paris: International Council of Museums: 566–73.
de la Rie, E.R. and C.W. McGlinchey. 1989. Stabilized dammar picture varnish. Studies in conservation 34(3): 137–46.
de la Rie, E.R. and C.W. McGlinchey. 1990a. New synthetic resins for picture varnishes. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the Contributions to the IIC Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 168–73.
de la Rie, E.R. and C.W. McGlinchey. 1990b. The Effect of a hindered amine light stabilizer on the aging of dammar and mastic varnish in an environment free of ultraviolet light. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the IIC Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 160–4.
de la Rie, E.R. and A.M. Shedrinsky. 1989. The Chemistry of ketone resins and the synthesis of a derivative with increased stability and flexibility. Studies in conservation 34(1):9–19.
Del Serra, A. 1985. A Conversation on painting techniques. Burlington magazine 127(982):4–16.
Dunkerton, J., J. Kirby, and R. White. 1990. Varnish and early Italian tempera painting. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the IIC Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 63–9.
Eastaugh, N. 1984. Gloss. The Conservator 8:10–14.
Feller, R.L.; N. Stolow, and E.H. Jones. 1985. On Picture varnishes and their solvents. Revised and enlarged ed. Washington, D.C.: National Gallery of Art.
Hansen, E.F., M.R. Derrick, M.R. Schilling, and R. Garcia. 1991. The Effects of solution application on some mechanical and physical properties of thermoplastic amorphous polymers used in conservation: Poly(vinyl acetate)s. Journal of the American Institute for Conservation 30(2):203–13.
Hoenniger, Cathleen. 1994. Early Italian paintings: Varnishing and aesthetics. In Varnishes: Authenticity and Permanence, L. Carlyle and J. Bourdeau, ed., Abstracts for the Colloquium, September 19–20, 1994, Ottawa, Canadian Conservation Institute: 12.
Kushel, D. 199?. Unpublished handouts to students. Buffalo State College, New York.
Lawrence, C. 1990. The Effects of solvent variations on long term properties of resin coatings. In Papers presented at the Sixteenth Annual Art Conservation Training Programs Conference. Winterthur, Del.: University of Delaware/Winterthur Museum Art Conservation Department: 65–82.
Peres, C. M. 1990. Vincent van Gogh's triptych of Trees in Blossom, Arles (1888) Part 2, On egg-white varnishes. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the ontributions to the IIC Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 131–3.
Richardson, J. 1983. Crimes Against the Cubists. New York Review of Books 30 (June 16, 1983):32–4.
Stoner, J.H. 1985. Ascertaining the artist's intent through discussion, documentation and careful observation. The International journal of museum management and curatorship 4(1):87–92.
Stringari, C. 1990. Vincent van Gogh's triptych of Trees in Blossom, Arles (1888), Part 1: Examination and treatment of the altered surface coatings. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the IIC Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 126–30.
Swaraj, P. 1985. Surface coatings: Science and technology. New York: John Wiley and Sons: 470–534.

Back to Paintings Chapter List