JAIC 2002, Volume 41, Number 1, Article 3 (pp. 21 to 42)
JAIC online
Journal of the American Institute for Conservation
JAIC 2002, Volume 41, Number 1, Article 3 (pp. 21 to 42)





Reginald Marsh, like Curry, began his career using oil paint in a straightforward manner. For instance, his 1928 New York Skyline (private collection) is unremarkable in technique. The canvas has a commercially applied oil ground, the paint behaves like pure oil paint that is not sensitive to solvents like acetone, and the surface has a slight gloss, although the painting is unvarnished.

But in 1929, the year after New York Skyline was painted, Marsh recorded in a notebook: “Met Tom Benton. He has boils on his neck and isn't painting. Comes up to my studio and shows me how to paint with eggs—fortunate event” (Marsh 1929). A few years later, Ralph Mayer recalled the experiments with tempera that both Marsh and Kenneth Hayes Miller (who taught at the Art Students League) carried out in 1929: “Marsh finally worked out a satisfactory formula with yolk of egg glazes as in watercolor but on white gesso.‘ Only this and nothing more' except a coat of varnish” (Mayer 1933). Marsh later described being introduced to egg tempera by Benton and by Denys Wortman (1887–1958). “It opened up a new world to me. Egg is a fine‘ draughtsman's' vehicle and very easy to handle. The luminosity and clearness of drawing is preserved, yet a certain greasy quality of the yolk gives a‘ fat' oily effect. Drying is instantaneous, and superimposed brush strokes are easily made. … I put egg yolk on a kind of belt line production for a dozen years and chucked oil forever.” Marsh said he used egg tempera from 1929 until he became dissatisfied with it in 1939 (Goodrich and Irvine 1955, 8, 13).

But with Marsh, as with Curry, the written evidence and the evidence of the paintings themselves indicate a much more complicated story.

Marsh had a lifelong habit of copying down information about technique. There are undated sheets in his papers containing notes from such early sources as Herringham's 1899 translation of Cennino Cennini and A. H. Church's late 19th-century book The Chemistry of Paints and Paintings (Marsh undated a). A notebook that can be dated between 1930 and 1937 is full of recipes for various kinds of egg-oil emulsions, including “Abell's” tempera (a mixture of whole egg, damar, linseed oil, and water), glue, gum, casein, and “Durex” media, as well as descriptions of glazing over tempera with pigments in varnish. The sources of these recipes include Max Doerner (via “Mrs. Koehl” in Munich), Kenneth Hayes Miller, Ralph Mayer, and Thomas Hart Benton. Notes that Marsh kept in 1931 indicate that he was experimenting with pigments mixed with gelatine or an unspecified gum for underpainting, followed by glazes in an egg medium (up to seven coats of “egg wash”); in 1932–33 he describes an “all egg picture” (Marsh 1930–37).

Examination of Marsh's paintings from these years gives further evidence that his description of the years 1929–39 as “egg yolk on a belt line” was an oversimplification. For example, Irving Place Burlesque (1930, William Benton Museum of Art, University of Connecticut) has some thick strokes of paint with bristlemarks that look exactly like oil paint straight from the tube. Some of the glazes on this painting are sensitive to solvents like acetone as well, which argues against a pure tempera medium. Similarly, the painting Two Women (1938, William Benton Museum of Art, University of Connecticut) has some areas of paint that are sensitive to solvents as mild as benzine.


Marsh once wrote,“Wouldn't it be wonderful to be an apprentice in the studio of Rubens?” (Marsh undated b). He must have thought that he had found the next best thing when he met Jacques Maroger, a former restorer at the Louvre, who had arrived in New York in 1940.

Maroger was not the first person who thought that he had discovered the secrets of the Old Masters. (For instance, a book called The Secret of the Old Masters had been published in 1906 by Albert Abendschein, a painter working in New York, and Abendschein's influence passed through Kenneth Hayes Miller to the generation that studied under Miller at the Art Students League [see Miller 1907 for Abendschein's influence on Miller, and Curry undated c and Marsh 1930–37 for recipes from Miller in those two painters' papers].) As discussed above, some of Maroger's ideas had appeared in English as early as 1935, and Vytlacil and Turnbull wrote that Maroger's media were being sold in America at that date (Vytlacil and Turnbull 1935, 66). But it was only upon his arrival in America that Maroger became very successful, attracting students in both New York and Baltimore, and part of the reason for this success may have been that Maroger found a devoted pupil and proselytizer in Reginald Marsh. In 1942, Marsh wrote a highly complimentary introduction to Maroger's first article in English, “The Secret of Van Eyck Regained” (Maroger 1942; Marsh 1942a).

Marsh had been ripe for a change in his technique. According to his own testimony, in 1939 (the year before he met Maroger) he had become dissatisfied with egg tempera, feeling that it lacked the richness of oil and that the darks were too opaque:“I'm through with tempera and egg yolk. It gets a painting all gummed up” (Goodrich and Irvine 1955, 13). Marsh was at that time experimenting with oil emulsions, watercolors, and Chinese ink (Goodrich and Irvine 1955).


What is most interesting for our purposes, because it documents so thoroughly the initial promise and subsequent controversy over Maroger's techniques, is the correspondence that soon developed between Marsh in New York and Curry in Wisconsin. Beginning in 1940, Marsh tried to convince John Steuart Curry that Maroger had, indeed, discovered the secrets of van Eyck, Rubens, and other Old Master painters, which had allegedly been lost since the death of Rubens.

Only some of Marsh's and Curry's letters are dated, but if the letters are arranged in a plausible sequence, it becomes clear that the correspondence was in full bloom by December 1940 and lasted at least until 1942. Marsh's earliest letters, which record Maroger's ideas at the time that he first arrived in America, are especially important because Maroger's theories were constantly evolving. The recipes that Maroger published in Magazine of Art in 1943 are different from those he advocated in 1940, and by the time he published his book The Secret Formulas and Techniques of the Masters in 1948, the recipes had changed again and had become still more elaborate (Maroger 1943, 1948).

In a 1940 letter, Marsh drew diagrams that introduced the various components of Maroger's system and explained to Curry how to use them (fig. 3). A 50% solution of gum arabic in water was component A in Marsh's diagram, and a solution of damar resin that had been dissolved by heating it with coldpressed raw linseed oil (one-third damar to two-thirds linseed oil) was component B. Marsh showed himself vigorously mixing equal portions of components A and B with a palette knife “like making mud pie” until it made an emulsion and stood up like mayonnaise. In one letter Marsh tells Curry to mix powdered pigments with this emulsion or “mayonnaise,” and then mix an additional 50% of the emulsion into the paints. In a second letter he contradicts this advice and says one should mix the dry pigments first with the damar-oil mixture, and then add the emulsion (Marsh 1940a, c).

Another element of the secret was something that Maroger called “black oil,” which was—in 1940—linseed oil cooked with 10% litharge by weight until it turned black. A sketch by Marsh (fig. 4) shows a palette set with the preground colors, a little pile of emulsion on the palette, and black oil in the palette cup (Marsh 1940b). Marsh told Curry to “smooch black oil & emulsion over the panel just prior to painting to get Rubens striping effect, draw in monotone and get black oil in every brush stroke—for opaque effects dip into black oil, for transparent effects into emulsion” (Marsh 1940a).

Fig. 3. Detail of letter from Reginald Marsh to John Steuart Curry [December 1940]. Archives of American Art, John Steuart Curry Papers

It is interesting that even at this date Devoe and Raynolds in New York were selling Maroger's black oil (which they called Devoe and Raynolds Special Dryer N10386) and the damar-oil mixture (which they called “Special White Dammar Varnish”). Marsh sent Curry a bottle of each of these on December 22, 1940, telling him that Devoe would not sell the materials to anyone except through the Maroger class at the New York School of Fine and Applied Art (Marsh 1940b).

Fig. 4. Detail of letter from Reginald Marsh to John Steuart Curry, December 22, 1940. Archives of American Art, John Steuart Curry Papers

In response to Marsh's descriptions of the Maroger method, Curry sent Marsh elaborate sheets of written questions, which Marsh then answered. McCloy acted as secretary in these exchanges, and sometimes transcribed Marsh's scribbled responses into neatly typed sheets for Curry. Marsh's responses clarified a number of points that were unclear in his initial letters, but they also document changes in the technique. For instance, an undated letter from Marsh was transcribed into typescript by McCloy under the heading “Addenda to Maroger's Method.” Maroger had apparently changed his mind and was at that time recommending that the colors be ground in black oil (not in the emulsion or damar-oil mixture as stated previously), except for the whites and yellows, which should be ground in linseed or walnut oil (Marsh undated c).

By January 1942, Curry had expressed his first doubts about the technique, asking whether any written documents could corroborate Maroger's views on Rubens's technique. A little more than a year earlier, Marsh had jokingly written to Curry: “Rubens used the Maroger medium—Rubens has just confirmed this from his grave—for you & me” (Marsh 1940d). But in response to Curry's serious question in 1942 about written corroboration, Marsh responded:“No, all is confused” (Curry 1942f).

In 1942, Curry turned his inquiries in a different direction, asking F. W. Weber of F. Weber Company in Philadelphia for his opinion of Maroger's materials. Weber warned Curry that the litharge in the black oil would make it tend toward yellowing and that gum arabic could make a film brittle. Weber made some samples of the Maroger media and sent them to Curry, and Curry then ordered larger quantities from Weber, saying that he would be glad to hear of any tests that Weber had carried out on the Maroger formulas. Curry must have been alarmed at the response he received from Weber on May 1, 1942: “We found that where the Gum Arabic was employed to obtain the emulsion, a very considerable lack of adhesion was noticed where the proportion of Gum Arabic in water was not kept down in quantity. … Even on an absorbent surface, where paint was employed impasto, the emulsion caused a breaking away and falling off of the paint” (Weber 1942a, b; Curry 1942ce).

The very next month (June 1942), a flurry of letters and transcribed questions and answers between Marsh and Curry shows that Maroger had introduced several changes to his method. Maroger was now recommending lead white instead of litharge to make his black oil. Marsh wrote that he had been working in Baltimore with Maroger, developing the new materials, and that Maroger started with 30% lead white, then changed to 20%, and then 10%, but had finally arrived at 3–4%. Maroger also introduced an entirely new medium, made of mastic varnish (either a solution of mastic in turpentine or mastic resin dissolved in linseed or walnut oil) mixed with the new black oil until it formed a jelly. Neither Marsh nor Maroger admitted it, but this was in fact megilp, a medium that had been known since the 18th century. Maroger was now recommending that the pigments on the palette be ground with linseed or walnut oil, not with an emulsion or with black oil (Marsh undated d, Marsh 1942b, c; on megilp, see Carlyle 2001, 101ff.).

Curry began to show signs of impatience. He wrote on June 17, 1942: “Dear Reggie: Thank you very much for your information. The new medium is as clear and transparent as a piece of coal.” Curry also asked whether the 10% litharge black oil that he had been using previously was dangerous and concluded: “Also, why has Maroger given up gum arabic? I thought that was the secret of Rubens. What does Rubens think now?” (Curry 1942g). Marsh's responses to these questions, if indeed he ever answered them, do not survive.


Paintings from this period by Maroger and by Curry show problems that may have led Maroger to make repeated changes to his recipes. A still-life painting by Maroger dated 1940 (private collection) has a fine pattern of traction crackle that is very disfiguring on a picture with such a smooth surface and such fine detail. A painting by Curry that was identified for us by McCloy as having been done using the Maroger media, Flowers (1940, private collection), shows very dramatic traction crackle as well as mechanical crackle and a serious lack of adhesion between paint layers that has caused flake losses in some areas.

McCloy showed us two paintings that he himself painted following Maroger's recipes at the time that Curry was corresponding with Marsh. McCloy's Horse in a Thunderstorm (1940, former collection of the artist, present location unknown) was painted following Maroger's first method—the one having gum arabic as an ingredient. The painting is generally in very good condition, with no crackle, although its gloss is very uneven. It does not look very much yellowed except a little in the most transparently painted areas (which presumably contain the highest percentage of medium). A Self-Portrait (private collection) by McCloy is undated, but McCloy remembered that it was done using Maroger's second medium (the one containing megilp). This painting, too, is in good condition and not noticeably yellowed. As is the case with the Doerner-type mixed technique, McCloy seems to have gotten good results using a method that had the potential to produce serious problems of preservation. Again, the reason for this success may have been that McCloy was a meticulous painter who allowed his layers to dry properly and refrained from using excess medium.

Marsh was not so fortunate, and many of his problems with the Maroger method seem the result of his having painted with too much medium. An exhibition label on the reverse of the painting Pursuit (1941–42, Toledo Museum of Art) has had the words “oil painting” crossed out and replaced by the artist with the words “Maroger medium.” This painting has a surface that is badly wrinkled from excess medium. Another painting that is annotated by Marsh is Seated Girl (1944, William Benton Museum of Art, University of Connecticut). Marsh wrote on the reverse of Seated Girl that he ground his pigments with black oil and mastic and also added a medium from his palette cup consisting of half stand oil and half turpentine. This painting, like Pursuit, has a surface that is very noticeably wrinkled. The artist's notations on the reverse of Seated Girl about the variations used on that particular painting are a caution against making generalizations about the exact materials used in “Maroger” paintings, even in a painting by one of the master's most devoted followers.

A portrait painted by Maroger himself, not dated but apparently from the 1940s (David Randall-MacIver, Lyman Allyn Art Museum), shows that Maroger himself had problems with excess medium. This painting does not have a discrete varnish layer but is glossy from medium that was mixed into the paint. There is some traction crackle and wrinkling, but the uneven yellowing of some of the lighter, cooler colors, caused by excess medium on the surface, is a very noticeable problem.

By 1948, when Maroger published his book The Secret Formulas and Techniques of the Masters, he had abandoned gum arabic completely but had developed many additional recipes that purportedly allowed one to paint like masters of various schools and periods. Egg yolk is added in some recipes, and beeswax is an ingredient in many others, especially in the recipes of Italian painters. Many unsubstantiated claims are made, such as the statement that the Dutch “Little Masters” did not use wax but Rembrandt used “the maximum quantity of wax” (Maroger 1948, 168). A detailed discussion of all of these recipes is beyond the scope of this article, but it is obvious that their increasing complexity is a further warning against making generalizations about paintings done using Maroger techniques.

In terms of solubility, the Maroger-type paintings done by Maroger, Curry, Marsh, and McCloy that we have tested often vary in solubility from area to area within each painting, which is what one might expect from a technique in which the artist was adding different media in various amounts directly on the palette. It is also interesting that the Maroger-type paintings that we have tested have been generally much less easily soluble than the Doerner-type paintings, probably because the Maroger recipes often involve adding a medium consisting of drying oil (the black oil), while many of the Doerner recipes involve adding easily soluble resins like damar. However, it should be noted that paintings made according to the later Maroger formulas described in his 1948 book (of which we have seen fewer examples) could potentially remain very easily soluble because many of the recipes contain wax.

Maroger media are still available from several sources, and the authors painted out test panels using materials ordered from David Davis in New York. It was impressive to see how effective the black oil is as a drier, making even very rich, oily paint mixtures dry to the touch overnight. This quick-drying quality could encourage a painter to paint with too much medium, running the risk of wrinkling and discoloration. Fine wrinkling was produced in samples where a substantial proportion of medium was added to commercially made oil paint, and after four years, the media used straight from the container, as well as light-colored paint mixtures containing a large proportion of medium, have turned noticeably yellow. After four years, samples of oil paint to which the wax medium had been added are still slightly soluble in benzine and are readily soluble in xylene. The tests suggested to us that one of the main reasons that painters have been attracted to the Maroger techniques (and tempera-oil methods as well) is that the quick-drying paint allows artists to apply multiple layers of paint, as earlier painters did, without having to wait long periods of time for the underlayers to dry.


Marsh's correspondence with Curry sheds some light on Marsh's views about varnishing. In January 1942, Curry asked Marsh whether Matvar (made by F. Weber Company) or wax could be used to dull the gloss if paintings done with the Maroger method were too shiny. Marsh responded that “beeswax can be put over after 2 weeks,” but added that “all the old masters were shiny” (Curry 1942f). In a sheet of lecture notes that Marsh wrote in 1941, a copy of which ended up in Curry's papers, Marsh said that if parts of a painting made with Maroger's methods were matte and other parts were glossy, the painting “can be varnished with a commercial varnish” (Marsh 1941).

Marsh's notebooks provide even more details about his varnishing practice and help to explain some of the peculiarities of his varnishes that have been noted by conservators. Marsh's paintings are often coated with thick, smooth varnish layers that have unusual solubilities and are susceptible to damage from abrasion, which can be annoying to conservators because such damage can be difficult to repair. The notebooks document the use of a great variety of different coatings, often applied in many layers on the same picture, including egg, “Mayer's damar,” “Montross mastic” (sometimes sprayed, sometimes brushed),“Toch's flat varnish,” and “Edgar Levy's wax.” Figure 5 shows Marsh's diagram of the eight layers of surface coatings that he applied in 1936 to his painting Bowery at Coney Island (1936, location unknown). Marsh was apparently dissatisfied with some of his varnish coatings, because several notebook entries say “varnish removed 1933—revarnished by Pichetto” (Marsh 1930–37).

Fig. 5. Detail from notebook of Reginald Marsh, 1936. Archives of American Art, Reginald Marsh Papers

Copyright 2002 American Institution for Conservation of Historic & Artistic Works