II. Traditional Artists' Varnishes

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
II. TRADITIONAL ARTISTS' VARNISHES


Author: Lance Mayer
Date: Submitted December, 1995
Compiler: Wendy Samet

TABLE OF CONTENTS:
A. INTRODUCTION
B. THE CLASSIFICATION OF NATURAL MATERIALS USED AS VARNISHES
C. DITERPENOID RESINS
D. TRITERPENOID RESINS
E. OTHER RESINS
F. NATURAL WAXES
G. GUMS AND GLUES
H. COATINGS DERIVED FROM EGG
I. RETOUCH VARNISH
J. COLORED VARNISHES
K. UNVARNISHED PAINTINGS

A. INTRODUCTION

1. Literary Evidence as a Source of Information about Traditional Varnishes

Artists' treatises and other written or verbal testimony from artists can give us information about varnishing practice, but this kind of evidence is scarce, especially for the earliest periods. Treatises have other limitations: they may not reflect typical practice but may feature “special” or exotic recipes, or may repeat earlier recipes that the author had not actually used (see Dunkerton et al. 1990, 66, for examples of recipes repeated for hundreds of years). Written information must also be used with caution because even if we can document that an artist used a certain kind of varnish at a certain point in time, many artists have changed practices over the course of their careers and might have used a different method on paintings of a different period (see Section K.2. below, Late 19th and Early 20th Centuries, on changes in the technique of Monet and Pissarro, for example).

2. Examination of Paintings as a Source of Information about Traditional Varnishes

a) Gross Examination of Paintings
Conservators often make rough judgments about the nature of varnishes on paintings based on the results of solvent tests, the examination of paintings' surfaces with a stereomicroscope, and the ultraviolet fluorescence of resin and oil-resin coatings. Conservators also look for stains through the crackle on the back of an unlined painting to judge whether a painting was cleaned and/or revarnished after crackle had developed, and therefore whether the existing varnish might or might not be the first varnish.
b) Scientific Analysis of Varnishes
(1) ANALYTICAL TECHNIQUES
The identification of aged varnishes by analytical means is difficult because the materials are chemically complex and because they degrade over time, making it difficult to compare samples with freshly-prepared standards. But progress has been made, especially by scientists at the National Gallery, London (see Mills and White 1977; Mills and White 1987; and White 1986), although most of the analyses of organic material that have been published are of paint media, rather than varnishes. New methods of analysis may yield further information in the future (Boon and Rainford 1994).
(2) CROSS-SECTIONS
The study of (under normal and ultraviolet light) and the staining of cross-sectional samples can also provide evidence about the nature and layer structure of varnishes. The interpenetration of a varnish layer into a paint layer, which has sometimes been reported (Wolbers 1990, 15, 31), may indicate that a varnish was applied while a paint layer was still very fresh, although varnish may also penetrate into an aged paint film that has been leached by previous cleaning. The discovery of a grime layer below a varnish layer (in cross sections, or during cleaning tests) may serve as evidence that the painting remained unvarnished for a considerable length of time before varnishing (Jones 1977, 10).

B. THE CLASSIFICATION OF NATURAL MATERIALS USED AS VARNISHES

Most 20th-century artists' manuals divide the natural resins used as varnishes into two groups: 1) Oil-resin varnishes, made from “hard resins,” such as the copals, which are usually heated in a drying oil to dissolve them, and which form tough insoluble films; and 2) Spirit varnishes, made from “soft resins” like damar and mastic, which dissolve in solvents like turpentine and which form relatively brittle films, but which remain soluble to some degree. This division is readily understood by conservators and by many artists, but unfortunately some historic varnishes do not fit very well into the scheme. Mills and White (1977; 1987) have outlined a classification of resins based on their chemical structure which is more complex but ultimately more accurate. In the following outline we will generally follow the Mills and White classification of resins into diterpenoids and triterpenoids. (Their publications should be consulted for more detailed information about the subclassifications and specific chemical structure of each material.)

C. DITERPENOID RESINS

This family includes a wide variety of tree resins, from conifer resins like Strasbourg turpentine, Venice turpentine, and sandarac, to the various kinds of copals.

1. Pine Resin (Colophony, Rosin, Violin Rosin)

This is the brittle, sticky resin obtained by distillation of various pine species (Pinus spp.). It appears as an ingredient in 14th-century Italian recipes (Dunkerton et al. 1990, 67 n.) as ragia di pino and pece Greca (see also Merrifield 1849 cclviii, ccli ff.), and in Armenini in the 16th century (Armenini [1586] 1977, 196). Both Doerner ([1922] 1949,106,134) and Mayer (1970, 196–7) assert that rosin is found as an adulterant in inferior 20th-century varnishes.

2. Venice Turpentine

The viscous, semiliquid secretion of the European larch (Larix decidua) appears frequently as an additive in early varnish recipes, although as Mills and White point out, it is not clear why, for it dries slowly to a brittle yellow film (Mills and White 1987,89). De Mayerne, in the 17th century, lists several recipes where Venice turpentine is either diluted with oil of turpentine as a retouch varnish, or added to mastic or sandarac varnishes, with or without oil (Talley 1981, 81, 126–27, 129–30). Reynolds is said to have used it as a final varnish, even over an egg varnish(!) (Talley 1986, 62–3). In the 19th century, Rembrandt Peale recommends it as a retouch varnish, much diluted (Peale ca. 1850, 194). Doerner is enthusiastic about Venice turpentine as a painting medium, but says it makes a poor final varnish (Doerner [1922] 1949, 126).

3. Strasbourg Turpentine (Olio di Abezzo)

This is a product of the European fir (Abies Alba). In recipes of the 16th century, it is sometimes used mixed with oil, but by the late 16th century Armenini gives a recipe for olio di abezzo in naphtha alone (Caley 1990, 70). In the 17th-century Paduan manuscript, it is mixed with naphtha and mastic (Merrifield 1849, 672).

The North American fir Abies balsama yields the similar Canada balsam. This is mentioned as a varnish by Rembrandt Peale, diluted with turpentine, but he says it is apt to chill or bloom, so considerable copal must be added (Peale ca. 1850, 95; 178). Sully ([1873] 1965, 24) says that Canada balsam makes a poor varnish because it grows dark.

4. Sandarac and Related Resins

Sandarac is one of the most important of the early varnish resins. Modern sandarac comes from the North African plant Tetraclinus articulata, but there has long been confusion between this and other similar resins of the Cupressaceae type which come from the juniper (Merrifield 1849, ccliii; Dunkerton et al. 1990, 68 n.). When heated with linseed oil, sandarac forms the vernice liquida which was apparently the most common early Italian varnish (Eastlake 1847, 225ff.); Caley says that from the 12th-15th centuries, “most of the varnishes in the literature are made from linseed oil and sandarac” (Caley 1990, 70).

Sandarac has recently been identified in a sample of a very early varnish from a 14th-century Italian painting. The varnish, consisting of sandarac, boiled linseed oil, and red lead (as a drier), is one of the few cases where scientific analysis has confirmed what we had previously known only from literary evidence (Dunkerton et al. 1990, 63–4). Armenini gives 16th-century recipes for sandarac as both an oil varnish and a spirit varnish (Armenini [1586] 1977, 196). By the 17th and 18th centuries, sandarac still appears, but more often as a spirit varnish (Merrifield 1849, 644; Talley 1981, 205; Veliz 1986, 187). Carlyle and Bourdeau give two 19th-century recipes (1994, 13–14). After the publication of Eastlake's book in 1847, Rembrandt Peale experimented with Eastlake's recipe for a linseed oil-sandarac varnish, which Peale tried as a painting medium (Peale ca. 1850, 86). Toch (1934, 150) says that sandarac in alcohol plasticized with castor oil or lavender oil is sometimes called “French varnish.”

5. The Copals

“Copal” is a general name for a number of hard resins from many different species of tropical trees, all of which dissolve in oil and produce tough, insoluble films. Mills and White divide them into Araucariaceae resins (which include Kauri resin and Manila copal), and Leguminosae resins (which include the African copals and South American copals). The specific resins are difficult to characterize by analysis, and the origin of some of the copals remains obscure (Mills and White 1987, 90–2).

Reynolds used copal (“from Birmingham”) as a painting medium (Talley 1986, 62), but the heyday of copal as a final varnish was the 19th century. Mérimée was a strong advocate of copal varnish; he believed it was stronger, and “preserves its original brightness” better than oil or mastic (Mérimée 1839, 91). Mérimée also recommended a system of layered varnishes, applying a thin layer of copal varnish followed by a layer of mastic which could be removed at a later date if it deteriorates (Mérimée 1839, 82, 91–2; see also Swicklik 1993, 160). Carlyle has found that this procedure is recommended in a number of 19th-century British treatises (Carlyle 1990, 79). In the United States, Rembrandt Peale quotes Field, who says that while mastic varnish is the best varnish for a painting that contains mastic added to the paint, copal is the best varnish for a painting in which copal has been used as a medium (Peale ca. 1850, 135). Sully says that “Some recommend copal varnish,” although he prefers mastic or damar (Sully [1873] 1965, 23). Church recommends copal as a final varnish (Church 1890, 105–6), and it was still widely used as a painting medium and an intermediate layer in the late 19th century (for instance, by Bouguereau Swicklik 1993, 161]), but its tendency to turn dark and insoluble had been noticed, and it is generally not recommended as a final picture varnish in the 20th century (Toch 1934, 149; Doerner [1922] 1949, 136–7).

6. Copaiba (Copaiva) Balsam

This is a liquid, related to the Leguminosae copals, that leaves a slight residue after evaporation. It appears occasionally as an additive in early varnish recipes (Merrifield 1849, cclxi), but its chief interest to conservators is that it was formerly used by conservators when cleaning paintings, in order to keep the surface “wetted up” after the cleaning solvent had evaporated (Mills and White 1987, 92; Ruhemann 1934, 7). It was also used to “feed” the paint as part of the discredited 19th-century Pettenkofer alcohol vapor treatment (Schmitt 1990).

D. TRITERPENOID RESINS

This group includes among others mastic and damar, the two most important natural resin ingredients in 19th- and 20th-century varnishes. (See also Section IV.B., Natural Resin Varnishes.)

1. Mastic

Mastic is derived from a species of Mediterranean shrub (Pistacia lentiscus), and for years its principal source has been the Greek island of Chios. Early recipes include mastic as an ingredient in oil-resin varnishes (Merrifield 1849, ccliv), and these recipes persist into the 16th and 17th centuries (Armenini [1586] 1977, 195–6; Merrifield 1849, 670). But this period also begins to see recipes for mastic with only a relatively small amount of oil added (Merrifield 1849, 632), as a spirit varnish dissolved in spirits of turpentine with only pine resin added (Talley 1981, 98–9), or dissolved in naphtha with olio di abezzo added (Merrifield 1849, 672). Mastic spirit varnishes were to become the most frequently recommended varnishes of the 19th century, and it is only in the second half of the 19th century (and even later in England) that damar begins to make inroads into its primacy (see discussion of damar, Section 2. below).

a) Additives
It was long recognized that while mastic has many advantages in terms of gloss, transparency, and removability, it is brittle and subject to degradation by moisture. Some 19th century recipes specify mastic alone dissolved in turpentine (Mérimée 1839, 80–1; Peale ca. 1850, 135), but resins and oils have been added to mastic in an attempt to plasticize or toughen it from the earliest times until the present day. These additives are important because they can help to explain the different properties that we find in varnish layers on old paintings. For example, some conservators feel that a higher-than-usual proportion of drying oil added to a spirit varnish can explain why some old varnishes are sticky under the cotton swab and are annoyingly difficult to thin evenly. Additives can also presumably account for difficulties in applying a modern varnish on top of residues of an old varnish, if there is a component in the old varnish that reacts unexpectedly to solvents.
Other additives, in addition to the ones cited above under D.1., include Venice turpentine (Peale ca. 1850, 177–8, citing de Mayerne), rosin (Mérimée 1839, 80–1), and Canada balsam, gum elemi, or Venice turpentine up to one-seventh the weight of mastic (Church 1890, 99). Sully tells his readers to add camphor to mastic varnish (Sully [1873] 1965, 22–3), and Church writes that in mastic varnishes from France camphor is added at the rate of 5–8% (Church 1890, 100). Toch (1934, 150) says that linseed or poppy oil “must be” added in small quantities to mastic varnish, but Doerner criticizes the practice although he says that it is often done by commercial varnish manufacturers (Doerner [1922] 1949, 130, 208).

2. The Damars

These are products of a large family of Asian trees (Dipterocarpoideae). The material called damar that is at present available from Malaysia and Indonesia “appears to be of fairly consistent composition though of imprecise botanical origin” (Mills and White 1987, 93). Damar was introduced more recently into Europe than mastic—its first recorded use as a varnish was in 1829, but for some time there was confusion about its identity (Feller 1966). Merrifield says in 1849 that damar varnish is much used in Venice and in Munich (Merrifield 1849, cclxi). Rembrandt Peale mentions damar (Peale ca. 1850, 95–6, 178), and speaks of its clarity and gloss, and its superiority to mastic. Sully prefers mastic, but describes damar as the best varnish, much used in the United States and “almost universally in Germany” (Sully [1873] 1965, 16, 23–4).

Britain seems to have lagged behind other countries in adopting damar. Church mentions it (1890, 53), but does not speak of its good qualities, and in his recipe for mastic varnish says only that damar, along with sandarac, “may be substituted wholly or in part” for mastic (Church 1890, 100). Carlyle has pointed out that damar varnish was not advertised by name in Britain even in the first decades of the 20th century (Carlyle 1994, 9), although it certainly could have been available earlier.

Substances have sometimes been added to damar to strengthen or plasticize it, as they have been added to mastic. Rembrandt Peale's recipe was simply damar in spirits of turpentine (Peale ca. 1850, 95–6), but in the 20th century Toch recommended adding up to 5% stand oil (1934, 151), and Doerner said that 5–10% castor oil can be added (Doerner [1922] 1949, 133).

3. Benzoin

Benzoin comes from species of trees (Styrax) that grow in Southeast Asia (Mills and White 1987, 95), but the material was imported into Europe early. It was one of the first spirit varnishes and appears in the early 16th-century Marciana manuscript, dissolved in spirits of wine (Merrifield 1849, 628–31). Armenini gives a benzoin spirit varnish recipe (Armenini [1586] 1977, 196) and benzoin varnishes are listed in the treatises of Pacheco (1649) and Palomino (1715–24) (Veliz 1986, 85, 187).

E. OTHER RESINS

Many other resins which may have been used occasionally as ingredients in varnish are not discussed here. Also not discussed are those resins used primarily as media rather than in surface coatings, like gum elemi (see Swicklik 1993, 161, for Bouguereau's use of gum elemi as a glazing medium). It is important to note that unusual ingredients or combinations of ingredients may well occur in varnishes, and in fact are likely to occur especially with the search for the “secrets” of the Old Masters and the proliferation of patent recipes in the 19th century. Even academically trained artists occasionally used varnishes not intended for paintings. Frank Duveneck, for example, is known to have used coach varnish (Mayer and Myers 1993, 136).

1. Amber

Amber is the fossilized resin of trees. The chemistry of amber is discussed by Mills and White (1987, 96–9). There are many old recipes for amber varnish, but there is some question about whether amber was distinguishable from hard copal resins, and whether such a valuable material as amber would have actually been used in a varnish (Merrifield 1849, ccliv-viii). A varnish called amber varnish was still sold by Winsor and Newton in 1913, but it has been pointed out that a true amber varnish would work poorly as a varnish for paintings (Toch 1934, 149–50). “It is known in the trade that only a very small quantity sold under this name really contains amber at all” (Hurst 1922, 409).

2. Shellac

Unlike the previously discussed resins which derive from exudations from plants, shellac is the secretion of an insect which is harvested primarily in India (Mills and White 1987, 101–3). Shellac is initially soluble in alcohol, and has been more often used as a coating for furniture and other objects than as a varnish for paintings. But Peale (ca. 1850, 183–4) quotes from Field's notes about “White Lac Varnish,” although Peale says he has not succeeded in obtaining a gloss with it. (Field's recipe is given in Carlyle and Bourdeau 1994, 16.) Surprisingly, shellac is also recommended in the 20th century by Toch (Toch 1934, 150) as a good picture varnish (bleached shellac in alcohol with oil of lavender spike added as a plasticizer). Doerner says that shellac should on no account be used as a picture varnish (Doerner [1922] 1949, 136).

3. Asphaltum and Bitumen

The transparent, brown tarry materials asphaltum and bitumen (see White 1986, 62–3) are mentioned here because they were apparently used to give a general toning to paintings by 19th-century artists (see, for instance, Sully [1873] 1965, 35, 37; see also Section J.2. below), and could thus be possibly found as an element in a sequence of varnish layers. Mérimée writes that when asphaltum is mixed with turpentine or drying oil “it then affords a very brown and transparent varnish” (Mérimée 1839, 46).

F. NATURAL WAXES

Beeswax occasionally appears as a final varnish, used especially by artists like Reynolds who were prone to experiment (Talley 1986, 62). Delacroix, who had great anxieties about varnish, had his varnisher try a wax varnish among other experiments (Swicklik 1993, 162). And later in the 19th century, Gauguin preferred a wax varnish on some of his paintings, which he said would preserve them from yellowing and from other deterioration, and which would presumably be more sympathetic to their matte surfaces as well (Christensen 1993, 92–3). Wax has been used in the 20th century to make natural resin varnishes more matte: either by mixing it directly into a spirit varnish (Toch 1934, 151); or by applying beeswax to an already dry varnish, as was recommended to conservators by Ruhemann in 1933 (Ruhemann 1968, 318–20). (See also Section VI., Wax as a Surface Coating.)

G. GUMS AND GLUES

Varnishes made from these materials are occasionally found. For instance, Cennini says that parchment size should be used as a preliminary coating in order to make terre verte take subsequent varnishing (Cennini 1960, 122). Carlyle and Bourdeau list a gum arabic recipe and an isinglass recipe taken from Dossie (1758) and an isinglass recipe from 1800 (Carlyle and Bourdeau 1994, 30–2). Delacroix's experiments included gelatine as a temporary varnish (Swicklik 1993, 162–3).

H. COATINGS DERIVED FROM EGG

1. History

Egg white (glair) varnishes have been used since the days of Cennino Cennini (Cennini 1960, 99–100). Caley points out that the Strasbourg manuscript in the 15th century mentions a varnish made with egg whites, gum arabic, and tree gum, while in the 17th century glair varnishes were criticized by de Mayerne and Sanderson (Caley 1990, 70–1). Carlyle and Bourdeau give several 18th- and 19th-century recipes which include additives like sugar, alcohol, brandy, garlic juice, and honey (Carlyle and Bourdeau 1994, 22–9). R. and P. Woudhuysen-Keller (1994) cover the topic exhaustively and publish written references to egg white varnishes dating from the 11th to the 20th centuries. The late 18th-century American painters Richard and William Jennys gave instructions on the reverse of a painting to varnish it each year with an egg white varnish (Wolbers 1988, 246). Egg white varnishes have been found on some of the paintings of van Gogh (Peres 1990).

2. Temporary Nature

It had long been understood that egg white varnishes were used chiefly as temporary varnishes which were intended to be removed when a painting had dried enough to apply a final varnish (Swicklik 1993, 163). But Reynolds is said to have interlayered egg varnishes with resin varnish layers and retouchings, and his contemporaries thought that this contributed to the deterioration of his paintings (Talley 1986, 62–3). Sully too, gives an egg varnish/sugar recipe which he says in one case he used as a temporary varnish; in another case he varnished over it with mastic ([1873] 1965, 23; 25–6).

3. Matteness

There are some hints that egg white varnishes may have been seen as providing a matte effect that was specifically desired by the artist. In the late 17th-century Volpato manuscript, the master asks his apprentice to varnish some of his paintings with mastic, but to coat others with white of egg, both during painting and after painting was complete (Merrifield 1849, 746, 748). More to the point, Talley quotes from Samuel Pepys' diary that Pepys saw a chimney piece by Hendrik Dankert “in distemper, with egg to keep off the glaring of the light, which I must have done for my room: and indeed it is pretty, but, I must confess, I do think it is not altogether so beautiful as the oil pictures; but I will have some of one and some of another” (Talley 1981, 242–3).

I. RETOUCH VARNISH

Some varnishes were considered especially useful for keeping colors from “sinking in” or becoming unevenly matte during the course of painting. Retouch varnishes are of special interest to conservators because their presence might pose solubility problems while cleaning a painting. Talley cites a recipe from de Mayerne for an oil-resin varnish that was supposedly used by painters in Italy to coat their dead colors and then paint into (Talley 1981, 144–5). Swicklik cites a lecture by Oudry, who recommends a thin layer of varnish over the sketch, and then another layer over the underpainting (Swicklik 1993, 158). Other sources indicate that this was not the usual practice among French painters of the time (Swicklik 1993, 158–9), although the late 17th-century Volpato manuscript discusses both “oiling out” and varnishing abozzi before continuing painting (Merrifield 1849, 746).

In the 19th century, French formulations for special retouch varnishes began to be cited. A retouch varnish made by Lahnee & Brother is mentioned as a new thing by Rembrandt Peak, supposedly much used in France and Germany, but by Peale only experimentally (Peale ca. 1850, 184). In the late 19th century, Vibert's retouch varnish, derived from damar with a little poppy oil, was available for many years (Mills and White 1987, 93). Many artists of the 19th century must have used unusual combinations of varnishes in their paintings in order to recreate the look of Old Master paintings, or to achieve enamel-like effects. For example, Bouguereau used mastic-oil varnish and copal varnish interlayered with his paint, as well as adding varnish to his body colors and glazes (Swicklik 1993, 161).

J. COLORED VARNISHES

1. The Color of Varnishes

Although artists' treatises show that painters often tried to obtain varnishes that were as clear as possible (An Exhibition of Cleaned Pictures 1947, iii-xxiv), the old techniques of cooking oil-resin varnishes sometimes resulted in varnishes that “are darker and warmer when applied, and deepen more with time than the modern soft resin spirit varnishes” (An Exhibition of Cleaned Pictures 1947, vii). When Rembrandt Peale experimented with Eastlake's recipe for linseed oil-sandarac varnish, he found it was dark red (Peale ca. 1850, 86), and this is what Dunkerton et al. (1990, 68–9) found when they recently tried to recreate sandarac-oil varnishes. In the 19th century, Mérimée says that many artists object to copal varnish because it has a slight color (Mérimée 1839, 82). Both Eastlake and Church thought that Italian tempera painters took the strong color of early varnishes into account when they painted their works, and Eastlake believed that they may have even deliberately tinted their varnishes (Eastlake 1847, 252, 270–1; Church 1890, 54).

2. Tinted Varnishes Applied by Artists

The degree to which Old Master painters may have applied tinted varnishes to their paintings or anticipated the darkening effects of age has been debated in a series of articles in The Burlington magazine (Brandi 1949, MacLaren and Werner 1950, Gombrich 1962 and 1963, Kurz 1962 and 1963, Mahon 1962, Plesters 1962).

By the 18th century, there is written evidence that Joshua Reynolds deliberately tinted some of his varnishes, either with pigments (Talley 1986, 67) or with asphaltum (Mérimée 1839, 334n). Reynolds is said to have passed this practice on to other English painters (An Exhibition of Cleaned Pictures 1947, xvii). Some researchers have found evidence of locally applied tinted varnishes on some of Reynolds's paintings, but no evidence of overall tinted varnishes on paintings examined to date (Bockrath and Buckley 1984, Buckley 1986). In America, Sully wrote that “toning the picture is a general practice” (Sully [1873] 1965, 37), and that Washington Allston “recommended me to use a very slight glazing or toning over every portrait I painted. Generally speaking, he thought asphaltum most fit for the purpose” (Sully [1873] 1965, 35; see also Stoner 1990). As in the case of Reynolds, it has been difficult to confirm this on actual American paintings (perhaps because the layers, if present, would be likely to have been removed during previous cleanings), although intermittent toning or “patinating” layers are sometimes found (Mayer and Myers 1988, 40).

3. Tinted Varnishes Applied by a Later Hand

It has been claimed that restorers of the 19th century very often applied a tinted varnish after having cleaned a painting (An Exhibition of Cleaned Pictures 1947, vii). This may have been done partly in the belief that some early artists used toned varnishes (see Section J.1. above). Church (1890, 278) writes that “Sometimes a little dragon's blood, or other warm-coloured resin, is added to the masticvarnish, in order to prevent the cold and raw look which a picture which has lost its old toned varnish frequently presents.”

But the practice must also have been influenced by the views of critics like Ruskin, who believed that “bold and frank” paintings like those by Rubens looked best when “mellowed by time into more perfect harmony,” and that they were best “seen under circumstances of obscurity,” even though Ruskin knew that Rubens had not intended them that way (An Exhibition of Cleaned Pictures 1947, ix). Later in the century, Monet suffered the indignity of having his paintings varnished by the dealer Durand-Ruel with bitumen in order to make them more salable (Bomford et al. 1980, 101; Swicklik 1993, 167).

K. UNVARNISHED PAINTINGS

1. Early Cases of Unvarnished Paintings

Certain 15th-century paintings, executed thinly on canvas using a distemper medium, are generally understood never to have been intended to be varnished (Wolfthal 1987). But Cennini indicates that even tempera paintings on panel were sometimes varnished and sometimes not (Cennini 1960, 122; see also Dunkerton et al. 1990; and Mayer 1977). In the 16th century, Vasari implies that oil paintings may not always have been varnished if their surfaces were already adequately saturated. For instance, in speaking of painting in oil on stone, he says that because the stone is not absorbent, these paintings “may or may not be varnished, just as you like” (Vasari 1960, 239). Concerning painting in oil on a wall, Vasari says that the artist can “finish such work in the same manner as he treats the panel, always having a little varnish mixed with the colours, because if he does this he need not varnish it afterward” (Vasari 1960, 232).

a) Deliberate matteness
The two references cited above from the Volpato manuscript and from Pepys' diary in Section H., Coatings Derived from Egg, imply that some kinds of paintings in the 17th century may not have been intended to be glossy. Even more to the point, Eastlake transcribes a passage from the 17th-century De Mayerne manuscript that describes a method to “make all kinds of colours sink in and look dull, and to prevent their shining,” by adding spike oil to linseed or nut oil (Eastlake 1847, 432). Eastlake cites similar recipes for making oil paint matte in several French treatises of the 17th and 18th centuries (Eastlake 1847, 432–4n).

2. Late 19th and Early 20th Centuries

The desire of some of the Impressionist and Post-Impressionist painters to have their paintings remain unvarnished is now well known. Swicklik found that artists differed in their approaches. For instance, Renoir seems to have consistently preferred his paintings varnished. Monet and Pissarro rejected varnish after 1880, but for different reasons: “Pissarro because of his desire for a matte finish; Monet lest it discolor his effects” (Swicklik 1993, 171). When Pissarro became a Neo-Impressionist, he adopted that movement's even more dogmatic opposition to varnishing (Swicklik 1993, 168–9). The Neo-Impressionists often exhibited their works behind glass in order to protect them because they were unvarnished (Herbert 1968, 21, 125). Gauguin's preference for wax as a substitute for varnishing has been discussed above in F., Natural Waxes. American Impressionist painters are not consistent. Some, like Childe Hassam, varnished some of their paintings, while John Twachtman can be documented as having wanted some of his paintings to look matte (Mayer and Myers 1993, 136–7).

Richardson writes that Picasso and Braque did not varnish their Cubist paintings and that conservators who have subsequently varnished those paintings have violated the intentions of the artists (Richardson 1983).

3. How Matte?

It should not be automatically assumed that unvarnished means matte. For instance, a waxed surface like Gauguin's beeswax varnish would have had a softer sheen than a mastic or damar varnish, but could probably not (when new) have been described as truly matte. As Swicklik (1993) points out, much of the impetus toward not varnishing was to prevent yellowing, rather than specifically to achieve a matte surface. However, by the first part of the 20th century, one does begin to find references to deliberate attempts to achieve a “dead, dull, lackluster, nontransparent look to the surface so much prized by some modern painters, who take special pains to bring it about,” and references to expedients like rubbing starch from a raw potato on a painting to eliminate glossy spots (Mayer and Myers 1993, 135, 137).

However, when unvarnished American Impressionist paintings are taken out of a frame having a wide rabbet, the parts that have been protected from light and air are often glossier than the exposed areas, leading to the suspicion that unvarnished paintings that now appear very matte may have been less matte when they were first painted. In fact, interviews with painters indicated that a number of American Impressionist painters added oil or other media to their paint, partly in order to get additional gloss so that the surface would be saturated without having to varnish it (Mayer and Myers 1993, 136). This idea is as old as Vasari (see K.1. above), and also appears in a British 19th-century treatise (Muckley 1893, 124–5 [Thanks to Leslie Carlyle for this reference]).

4. Interpreting the Evidence of Varnished and Unvarnished Paintings

a) If a painting has come down to us unvarnished, weight is usually given to this fact in determining whether this was the artist's intent. However, even here there are caveats: the artist may have considered the painting unfinished or may have intended to wait for the recommended period of six months to a year before varnishing, by which time the painting may have been out of the artist's hands.
b) Conversely, a painting which can be shown to have been varnished shortly after it was completed could conceivably have been varnished by a dealer or owner against the artist's wishes. In fact, it can be documented that artists from the 17th to the 19th centuries often understood that the varnishing of their paintings would be done by someone other than themselves (Swicklik 1993, 163–5). Therefore, it may be more useful to abandon the expression “original varnish” and replace it with the term “first varnish.”
If a conservator can establish that there is artist's retouching on top of a varnish layer, this is one of the rare cases in which there is little question that the painting was varnished while it was still in the artist's hands.
c) Because of the above uncertainties, conservators sometimes make subjective aesthetic judgments about the appropriateness of varnish on a painting. For example, some paintings by the American Impressionist Theodore Robinson have remained unvarnished, while paintings by Robinson that have been varnished sometimes show dark strokes of paint that look aberrant and do not conform to the modeling of the rest of the painting. When the varnish is removed and the surface is more matte, these dark strokes become more harmonious and it seems a good guess that this is the way that Robinson intended them. (In this case additional weight is given to the argument by the fact that Robinson was a close associate of Monet, who preferred some of his paintings unvarnished.)
d) In the case of a living artist, the conservator can of course consult the artist. Conservators can also change an artist's mind. Georgia O'Keeffe was influenced by contact with conservators and after 1947 expressed a preference for thin coatings of sprayed synthetic varnish, although she had previously not varnished her paintings (Van Vooren 1994).

REFERENCES

[Anon.]. 1947. An Exhibition of cleaned pictures (1936–1947). London: The National Gallery.
Armenini, G.B. 1977 [first published 1586]. On the true precepts of the art of painting. E.J. Olszewski, transl. and ed. New York: Burt Franklin & Co., Inc.
Bockrath, M. and B.A. Buckley. 1984. An Investigation of the materials and techniques of Sir Joshua Reynolds. Talk given at the Paintings Specialty Group at the twelfth annual meeting of the American Institute for Conservation, Los Angeles, California, May 19, 1984.
Bomford, D., J. Kirby, J. Leighton, and A. Roy, with contributions by R. White and L. Williams. 1990. Art in the making: Impressionism. London: National Gallery Publications.
Boon, J.J. and D. Rainford. 1994. The Mass spectrometric identification of natural resins and their degradation products in and on paintings. Talk given at the colloquium, Varnishes: Authenticity and permanence. Ottawa: Canadian Conservation Institute, 19–20 September 1994.
Brandi, C. 1949. The cleaning of pictures in relation to patina, varnish and glazes. Burlington magazine 91(556):183–8.
Buckley, B.A. 1986. Sir Joshua Reynolds, The Ladies Annabel and Mary Jemima Yorke. Cleveland Museum of Art Bulletin 73(9)(November):350–71.
Caley, T. 1990. Aspects of varnishes and the cleaning of oil paintings before 1700. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works:70–2.
Carlyle, L.A. 1990. British nineteenth-century oil painting instruction books: A survey of their recommendations for vehicles, varnishes and methods of paint application. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works: 76–80.
Carlyle, L.A. 1991. A Critical analysis of artists' handbooks, manuals and treatises on oil painting published in Britain between 1800–1900: With reference to selected eighteenth century sources. Ph.D. dissertation, Courtauld Institute of Art, University of London. 2 vol.
Carlyle, L.A. 1994. Reproducing traditional varnishes: Problems in representing authentic surfaces for oil paintings. Talk given at the colloquium Varnishes: authenticity and permanence. Ottawa: Canadian Conservation Institute, 19–20 September 1994.
Carlyle, L. and J. Bourdeau. 1994. Varnishes: Authenticity and permanence. Workshop Handbook. Ottawa: Canadian Conservation Institute.
Cennini, C. d'A. 1960 [reprint of 1933 ed.]. The Craftsman's handbook, translation of the Italian “Il libro dell'arte.” D.V. Thompson, Jr., transl. New York: Dover Publications.
Christensen, C. 1993. The Painting materials and technique of Paul Gauguin. In Conservation research (Studies in the history of art no. 41, Monograph Series II.) Washington: National Gallery of Art:63–103.
Church, A.H. 1890. The Chemistry of paints and painting. London: Seeley and Co. Limited.
Doerner, M. 1949 [originally published in German, 1922, first American translation, 1934]. The Materials of the artist and their use in painting with notes on the techniques of the old masters. Revised ed. E. Neuhaus, transl. New York: Harcourt, Brace and Company.
Dunkerton, J., J. Kirby and R. White. 1990. Varnish and early Italian tempera paintings. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works:63–9.
Eastlake, C.L. 1847. Materials for a history of oil painting. London: Longman, Brown, Green and Longmans.
Feller, R.L. 1966. First description of dammar picture varnish translated. Bulletin of the American Group: International Institute for the Conservation of Historic and Artistic Works 7(1):8, 20.
Gettens, R.J. and G.L. Stout. 1966 [republication of 1942 ed.]. Painting materials: A short encyclopedia. New York: Dover Publications, Inc.
Gombrich, E.H. 1962. Dark varnishes: Variations on a theme from Pliny. Burlington magazine 104:51–5.
Gombrich, E.H. 1963. Controversial methods and methods of controversy. Burlington magazine 105:90–3.
Herbert, R.L. 1968. Neo-Impressionism. Princeton, N.J.: Van Nostrand.
Hurst, G.S. 1922. A manual of painters' colours, oils, and varnishes. Sixth ed., revised by N. Heaton. London: Charles Griffin & Company, Limited.
Jones, E.H. 1977. Monet observed. In Monet unveiled: A new look at Boston's paintings. Boston: Museum of Fine Arts:6–10.
Kurz, O. 1962. Varnishes, tinted varnishes, and patina. Burlington magazine 104:56–9
Kurz, O. 1963. Time the painter. Burlington magazine 105:94–7.
Maclaren, N. and A. Werner. 1950. Some factual observations on varnishes and glazes. Burlington magazine 92:189–92.
Mahon, D. 1962. Miscellanea for the cleaning controversy. Burlington magazine 104:460–70.
Mayer, L. 1977. The examination and treatment of an unvarnished painted crucifix by Neri di Bicci. In Student papers presented at the third annual art conservation training programmes conference, May 8th to 11th, 1977. Queen's University, Kingston, Ontario. 61–7.
Mayer, L. and G. Myers. 1988. Thoughts on the appearance of nineteenth-century American paintings. In Paintings Specialty Group of the American Institute for Conservation: Postprints of the papers presented at the sixteenth annual meeting, New Orleans, Louisiana, June 1–5, 1988. Washington, D.C.:American Institute for Conservation of Historic and Artistic Works:39–44.
Mayer, L. and G. Myers. 1993. Understanding the techniques of American tonalist and Impressionist painters. Journal of the American Institute for Conservation 32(2):129–39.
Mayer, R. 1970. The Artist's handbook of materials and techniques. 3d ed. New York: Viking Press.
Mérimée, M.J.-.F.-.L-. 1839. The Art of painting in oil, and in fresco: Being a history of the various processes and materials employed, from its discovery by Hubert and Jan van Eyck to the present time. Trans. W.B. Sarsfield Taylor. London: Whittaker & Co.
Merrifield, M.P. 1967 [republication of 1849 ed.] Original treatises on the arts of painting. New York: Dover Publications.
Mills, J.S. and R. White. 1977. Natural resins of art and archaeology. Their sources, chemistry, and identification. Studies in conservation 22(1):12–31.
Mills, J.S. and R. White. 1987. The Organic chemistry of museum objects. Sevenoaks: Butterworths.
Muckley, W.J. 1893. A Handbook for painters and art students on the character and use of colours. 4th ed. London: Baillicre, Tindall, and Cox.
Peale, R. ca. 1850. Notes of the painting room. Manuscript in Historical Society of Pennsylvania. Because the original manuscript has inconsistent pagination, page numbers given here follow those of a transcription made by Carol Hevner, a copy of which was kindly provided by Ross Merrill.
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 contributions to the Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works:131–3.
Plesters, J. 1962. Dark varnishes: Some further comments. Burlington magazine 104:452–60.
Richardson, J. 1983. Crimes against the Cubists. The New York Review of Books 30 (June 16):2–4.
Ruhemann, H. 1934. A Record of restoration. Technical studies in the field of fine arts 3:3–15.
Schmitt, S. 1990. Examination of paintings treated by Pettenkofer's process. In Cleaning, retouching and coatings: Technology and practice for easel paintings and polychrome sculpture. Preprints of the contributions to the Brussels Congress, 3–7 September 1990. London: International Institute for Conservation of Historic and Artistic Works:81–4.
Stoner, J. 1990. Washington Allston: Poems, veils and “Titian's dirt.” Journal of the American Institute for Conservation 29:1–12.
Sully, T. 1965 [originally published 1873]. Hints to young painters. New York: Reinhold Publishing Corporation.
Swicklik, M. 1993. French painting and the use of varnish, 1750–1900. In Conservation research (Studies in the history of art, no. 41, Monograph Series II.) Washington, D.C.: National Gallery of Art: 157–74.
Talley, M.K., Jr. 1981. Portrait painting in England: Studies in the technical literature before 1700. (Privately printed) Guildford: Paul Mellon Centre for Studies in British Art.
Talley, M.K., Jr. 1986. “All good pictures crack”: Sir Joshua Reynolds's practice and studio. In Reynolds. N. Penny, ed. London: Royal Academy of Arts:55–70.
Toch, M. 1934. Dammar as a picture varnish. Technical studies in the field of fine arts 2:149–51.
Vasari, G. 1960 [republication of 1907 edition]. Vasari on technique. 7th ed. L.S. Maclehose, transl. New York: Dover Publications.
Veliz, Z. 1986. Artists' techniques in Golden Age Spain: Six treatises in translation. London: Cambridge University Press.
Van Vooren, C. 1994. O'Keeffe's ‘Cross with Red Heart’: Alteration of the shape of a painting. In AIC Paintings Specialty Group postprints 1994: Papers presented at the 22nd annual meeting of the AIC, Nashville, Tennessee, June 10, 1994. Washington, D.C.:American Institute for Conservation:61–7.
Watin, J.F. 1744. L'Art du peintre, doreur, vernisseur. 1st ed. Paris.
White, R. 1986. Brown and black organic glazes, pigments and paints. National Gallery Technical Bulletin 10:58–71.
Wolbers, R.C. 1988. Aspects of the examination and cleaning of two portraits by Richard and William Jennys. In American Institute for Conservation of Historic and Artistic Works: Preprints of papers presented at the sixteenth annual meeting, New Orleans, Louisiana. Washington, D.C.:American Institute for Conservation of Historic and Artistic Works: 245–60.
Wolbers, R.C., with N.T. Sterman and C. Stavroudis. 1990. Notes for workshop on new methods in the cleaning of paintings. Getty Conservation Institute, August 1990.
Wolfthal, D. 1987. The technique of early Netherlandish canvas painting: An overview. In ICOM Committee for Conservation: 8th Triennial Meeting, Sydney, Australia, 6–11 September, 1987. Preprints, Vol. 1. Los Angeles: Getty Conservation Institute:119–22.
Woudhuysen-Keller, R. and P. Woudhuysen-Keller. 1994. The History of eggwhite varnishes. Bulletin, The Hamilton Kerr Institute (2):90–141.



Back to Paintings Chapter List