4 VARIATIONS IN GLOSS AND SATURATION ON UNVARNISHED PAINTINGS

An unvarnished painting is not necessarily completely, uniformly matte. Unvarnished American Impressionist paintings vary greatly in gloss one from another, and often a painting will vary in gloss from one area to another within the painting. Unvarnished paintings by French Impressionist and Post-Impressionist painters also show variations of gloss, both from painting to painting and within an individual painting. For example, paintings by Monet that have never been varnished can sometimes have areas that are quite matte, while other areas of the same painting have a low sheen.

An artist can control the gloss in his or her painting to some degree by choosing an absorbent ground that will absorb medium from the paint and make paint more matte, or by adding medium to the paint as it comes from the tube (as some American Impressionists were said to have done), which would tend to make it more glossy. But some of the variations in gloss within a painting may be accidental, depending upon the amount of oil that was mixed with the pigment in each tube of paint by the paint manufacturer. In unvarnished paintings by Monet, another effect appears to be at work. The more thinly painted areas tend to be more matte, as if the ground drew some of the medium out of the lower paint layers, but in more thickly painted areas, where multiple layers of paint are present, this effect is less pronounced, and these areas tend to be somewhat glossier. If Monet placed a quantity of paint on a blotter to remove excess oil, as eyewitnesses described, this step could also cause variations in saturation because the paint that was closer to the blotter would presumably be more lean.

The variation of gloss from one part of a painting to another is important because each area in a painting may appear lighter or darker in value, or weaker or stronger in intensity of color, depending upon its degree of saturation. Dark areas of paint, especially some dark blues, can appear light in value if they are unsaturated and much darker in value if they are saturated. Gloss and saturation are clearly connected, but differences of value and intensity caused by differences of saturation seem at least as important as issues of gloss or matteness per se.

If parts of a painting are less saturated than other parts—even if this effect is partly accidental—and the painting is then varnished, the relationships in the design may be thrown off because the less-saturated areas will become darker (and perhaps more intense in color) than they were when the artist saw the painting in its unvarnished state on his or her easel. Even a painting having a uniformly unsaturated surface could be affected in unexpected ways by varnishing, because pigments are changed in different ways by increased saturation. As will be shown in the discussion of Willard Leroy Metcalf (1858–1925) below (sec. 6.5), there is evidence that some American painters may have feared the changes in value that would be the result of a painting's being varnished against an artist's wishes. The 1898 comment by Daniel Burleigh Parkhurst cited above (sec. 2) also relates directly to this discussion. Parkhurst wrote that if a painter is intending to not varnish and have a painting remain matte, the painter should use stronger, more “pure” tones than he or she intends, because the matte surface will make the colors look less intense. The converse of this argument is that if an artist intended a painting to be matte, and someone later varnished it against the artist's wishes, some color areas might as a result be made too intense, which is exactly what may occur in some paintings by Theodore Robinson (discussed below in sec. 6.6).

In the past, conservators have sometimes applied a varnish having a matte surface (obtained by spraying or by adding matting agents to the varnish itself) if it was felt that an artist intended a painting to remain unvarnished. A matte surface would give a varnish layer a reflectance somewhat closer to that of an unvarnished paint surface. However, the uniform saturation of the paint by the varnish negates any differences of saturation (and hence variations of value and color intensity) that the artist saw on the easel and could therefore have an adverse effect on the way the parts of a painting relate to one another.

Allowing a painting to remain unvarnished or removing a varnish that had been inappropriately applied can help to reestablish the variations of saturation and gloss that can have such an important effect on the values and colors in different parts of the design and the ways in which those different parts interact to produce the illusion of three-dimensional space. Of course, the decision on the part of a conservator to thin or remove a varnish from an American Impressionist painting must be made on a case-by-case basis. In some instances, the solubilities of the paint or the varnish can make it impossible to reduce a varnish layer to the degree that the artist's variations of gloss become visible without risking damage to the paint. In many cases, however, it is possible to safely thin or remove a varnish and reclaim to some degree its variations of saturation and gloss.

Another complication in this discussion is that a painting's gloss can change over time. This change is difficult to measure precisely, but conservators sometimes observe in paintings whose edges have been covered by a wide frame rabbet that the parts of the painting that have been exposed to light and air are much more matte. Such cases are a caution against prizing matteness too much, because we might be admiring an effect caused by the passage of time rather than something that the artist intended. It must also be admitted that—in theory at least—different parts of a painting might change in gloss over time to different degrees. This type of change could happen, for instance, if various areas of paint were mixed with media that aged differently, such as resins that would degrade faster than oils when exposed to light and air, or if different pigments aged differently over time.