OLD MASTER RECIPES IN THE 1920s, 1930s, AND 1940s: CURRY, MARSH, DOERNER, AND MAROGER
LANCE MAYER, & GAY MYERS
2 JOHN STEUART CURRY
2.1 THE DEVELOPMENT OF CURRY'S TECHNIQUE PRIOR TO 1932
Curry's earliest paintings appear to have been done with a straightforward oil technique. Curry kept a notebook in which he documented the materials used in many of his paintings, and the earliest paintings are described in the notebook simply as “oil.” We were able to examine Kansas Pastures (1930, private collection), which is so described in the notebook (Curry 1932–38, 12), and saw that it indeed has the appearance and solubilities of a pure oil painting. (In this and other cases, we carried out small solvent tests, primarily to detect resinous additives; ready solubility in solvents such as acetone was taken to indicate the addition of a significant amount of resin to the paint.) Some parts of Kansas Pastures are quite glossy, as if a considerable amount of drying oil had been mixed into the paint in those areas, but when the painting was studied in 1997, it was found to be very well preserved, with little traction crackle or mechanical crackle and no flake losses. Curry had, however, noted technical problems with some of his early paintings, such as the 1921 Deer in Rocquette Lake (location unknown), described in his notebook as “oil.” He wrote of this painting: “Underpainting had begun to show through at time of sale 1930” (Curry 1932–38, 3). And of the 1929 Prayer for Grace (Jean Chapman Born; also described as “oil”), he recorded in his notebook:“This picture has begun to darken and crack. Painted on a grey ground which never seemed properly cured” (Curry 1932–38, 12). The problems that Curry noticed on paintings like these may have resulted, in part, from the tendency of oil paint to become more transparent over time (which could reveal a dark underlayer more clearly) or the fact that oil paint itself can become darker, especially if the percentage of medium is high (as it was in Kansas Pastures).
The development of such flaws may have contributed to Curry's dissatisfaction with his technique and his search for more stable materials. Curry had read Martin Fischer's The Permanent Palette (1930) shortly after its publication, and he wrote to Fischer asking for more detailed information about some materials described in the book. Fischer's response, dated May 8, 1931, included advice on how to obtain materials, including Valspar spar varnish, “Flugina” (a German boiled linseed oil), and Dutch stand oil, all of which were recommended in The Permanent Palette in place of spirit varnishes (Fischer 1931). The correspondence between Curry and Fischer marks the beginning of two themes that will continue throughout Curry's career: the artist's interest in obtaining detailed technical information and the varying quality of the information that he received. Coating paintings with oil or oil-resin varnishes, as Fischer recommended, could have put them at risk of irreversible discoloration.
2.2 DOERNER'S RECIPES AND THE TEMPERA REVIVAL
John Steuart Curry's The Line Storm (1934, Babcock Galleries) was considered one of the artist's most important paintings during his lifetime (fig. 1). For many years, the painting was thought to have been lost, and when The Line Storm was rediscovered several years ago, it was brought to the authors for conservation treatment. According to Curry's biographer, Laurence Schmeckebier (1943), The Line Storm was one of the first works that Curry painted using tempera underpainting and oil-resin glazes after having read Max Doerner's The Materials of the Artist. The occasion of treating the painting seemed a perfect opportunity to see whether written evidence or information from other sources could corroborate or contradict Schmeckebier's assertion about the influence of Doerner's book.
Conservators may know Doerner's The Materials of the Artist and Their Use in Painting, with Notes on the Techniques of the Old Masters (to give its full title in English) partly from the very negative review written by Helmut Ruhemann in the 1960s. Ruhemann pointed out that copies of old paintings made by Doerner's students had discolored and remained easily soluble, and he concluded (correctly, as we now know) that Doerner had made many wrong guesses about the techniques of earlier painters (Ruhemann 1964; see, for example, Bomford et al. 1988 on Rembrandt's medium). But it is important to remember that Doerner's book was received very differently when it was first translated into English in 1934. George Stout's generally very complimentary review in Technical Studies pointed out “slight flaws and omissions”—for instance, Stout was more skeptical than Doerner about the wisdom of adding resin or wax to oil paint. But Stout concluded by calling the book a “splendid labor” and wrote that “the art of painting owes [Doerner] a large debt” (Stout 1935, 46, 50). According to William McCloy, the reaction of painters working in the 1930s was also very favorable, because the book was the first one of its kind to have been written by a practical painter. McCloy told us that Doerner's book included so much useful information that it was soon considered “the artist's bible” (this and subsequent McCloy quotations are personal communication 1997).
John Steuart Curry, The Line Storm, 1934, tempera and oil on canvas, 76.3 x 121.9 cm (30 x 48 in.). Babcock Galleries, NewYork
Max Doerner's book, which was first published in German in 1921, can be seen as part of a larger revival of tempera painting during the first part of the 20th century. In 1901, Christiana Herringham (who had published an annotated translation of Cennino Cennini's Libro dell'Arte in 1899) founded in London the Society of Painters in Tempera, which published occasional technical papers by its members. The same year—1901—saw the publication in Germany of Ernst Berger's theory that Van Eyck used egg-oil emulsions (Berger 1901), and in France, J. G. Vibert published a book at about this time on painting in egg tempera in the manner of “primitive” painters (Vibert ca. 1900). By the 1920s the tempera movement had clearly gathered steam—A. P. Laurie's 1926 book The Painter's Methods & Materials documents the revival of painting with egg yolk and emulsions at that time in Britain and mentions the proliferating varieties of tempera paints being sold by colormen (Laurie 1926). There was sufficient demand by 1928 to justify the revision and reprinting of the earlier Papers of the Society of Painters in Tempera ([1901–7] 1928).
The tempera revival had crossed the Atlantic by the 1920s. The American painter Thomas Hart Benton (1889–1975), who had taught himself to paint with egg tempera in 1925–26 (Benton 1969, 57–58), showed Reginald Marsh how to use egg tempera in 1929 (Marsh 1929). John Sloan (1871–1951) had begun using a kind of tempera (which he called “flaxseed emulsion”) for underpainting in 1928 (Sloan with Farr 1939, 299, 300). Curry owned a copy of Laurie's 1926Painter's Methods & Materials, which cited the earlier publications by Berger in Germany and the Society of Painters in Tempera in Britain (Laurie 1926, 190), but which did not mention the German edition of Doerner's book. (In a recent article, Sprague  describes possible routes for the transmission of information about tempera from Britain to America via the British painter Maxwell Armfield [1881–1972], who was in the United States from 1915 to 1922, and Edwin Waldo Forbes [1873–1969], who taught courses on technique at Harvard beginning in the 1920s. However, we have not seen any evidence that either of these men had a direct influence on the artists we have studied.)
Many of Doerner's recipes involve a “mixed technique” of underpainting with an egg-oil emulsion, followed by oil-resin glazes. When we looked into Schmeckebier's allegation that Curry had used these kinds of materials on The Line Storm after Doerner's book was published in English in 1934, we found that the situation was more complex than it had appeared at first.
We found, first of all, that Curry (and Reginald Marsh as well) had obtained some of Doerner's recipes from sources other than Doerner's book. An undated manuscript sheet in Curry's papers, headed “Doerner Technique according to student in Munich,” describes grounds made from glue, zinc white, and whiting; oil-tempera underpainting; and overpainting in oil-resin colors (Curry undated a). Curry's 1932–38 notebook also lists several recipes attributed to Doerner, including a detailed recipe for a glue/zinc white/chalk/oil ground called “Prof. Dörner method for preparing chalkground canvas— obtained from Emil Gauss” (Curry 1932–38, 98ff.), and egg-oil emulsion and glazing media “from Bocour Co. after Dörner” (Curry 1932–38, 101ff.). A notebook kept by Reginald Marsh describes an eggoil emulsion recipe under: “Doerner—Mrs. Koehl, Munich” (Marsh 1930–37). Although none of these recipes are dated, they may well have been obtained—in fact, are likely to have been obtained— prior to 1934, when the publication of The Materials of the Artist in English made such recipes easily available in America without the aid of intermediaries.
In fact, the first firmly dated reference to a “mixed” technique in Curry's notebook occurs in 1932, two years before Doerner's book appeared in English: “Began using tempera underpainting on Circus material” (Curry 1932–38, 16). The descriptions of many of Curry's circus paintings are terse, most being listed simply as “tempera-oil,” but some of the 1932 “tempera-oil” paintings have more detailed descriptions that describe glazes containing 50% damar and 50% oil (Curry 1932–38, 17), as in one of the Doerner recipes that also appears in Curry's notebook (Curry 1932–38, 101). Some of the 1932 circus paintings are described in Curry's notebook as being painted on a “chalk ground”— quite possibly the Doerner recipe obtained from Emil Gauss that is cited in the same notebook—and one includes Venice turpentine as an ingredient in a painting medium (Curry 1932–38, 19), which again is a Doerner favorite cited in Curry's notebook (Curry 1932–38, 99–100).
A little later, Curry's descriptions of techniques become more detailed for some paintings. The artist's notes on The Line Storm (painted in November 1934) read: “tempera-egg underpainting—white shellac in alcohol over tempera—Stand oil and turpentine for painting medium for oil overpainting.” The previous entry, for Portrait of Sue Marsh (location unknown), begun April 20, 1933, describes “Dorner chalk ground,” followed by an egg-oil emulsion, coated with a damar–stand oil mixture. The following entry, for The Fugitive (private collection), begun in November 1934, shows still other differences: “Nearly full tempera underpainting—egg yolk medium—Damar and mastic varnishes between— Stand oil thinned with turpentine for oil finish.” In the winter of 1935–36 The Fugitive was “Lightened with tempera white and repainted with Dammar & oil glazes” (Curry 1932–38, 96).
When seen in context, The Line Storm turns out to be less a revolutionary technical departure than part of a sequence of many variations that Curry was trying out before and after 1934. Instead of following a single recipe, each painting was painted a little differently, as if Curry was trying to learn from his experiments. Occasionally, during these years, we find comments like:“Found ground very absorbent” or “This did not dry properly” (Curry 1932–38, 20, 96). It is also significant that, interspersed with recipes attributable to Doerner, Curry's notebook lists recipes from many other sources, including Harold Zimmerman, James Watrous, Helen Farr, and Lynn Fawcett (Curry 1932–38, 99, 102, 103, 104). Kathleen Curry remembered that about this time, she worried that her husband was going to poison them by using her kitchen pots and pans for experimenting with painting materials (personal communication 1997).
2.3 PAINTINGS BY CURRY INFLUENCED BY DOERNER, AND PROBLEMS OF PRESERVATION
The Line Storm shows some of the problems of preservation that will become very familiar in Curry's paintings done with a mixed tempera-oilresin technique. One problem is traction crackle. In the case of The Line Storm, the shellac layer that he described putting over his tempera underpainting may have been too glossy for the upper layers of paint to stick to, although cracking may have also been exacerbated—in this and other cases—by the practice of applying faster-drying, resin-containing layers on top of slower-drying ones.
Many paintings by Curry from around this time that we have examined or treated show similar problems. The Rainbow (Babcock Galleries) is undated but probably dates to around 1935, when Curry was painting other versions of the subject (Curry 1932–38, 21). While The Rainbow is not described in his notebook, Curry made pencil annotations on the unpainted margin of the panel itself, which read: “Shellac size/Aug 13th/Repainted over wet paint.” In this painting, some areas of wide traction crackle are found where fairly matte passages of oil paint are clearly not adhering to the glossy shellac underlayer. Curry's inscription hints that the timing of the application of paint over layers that were not dry may also have been a factor and corroborates McCloy's description of Curry as an impatient, impulsive painter. The Rainbow also demonstrates the ready solubility of many of Curry's paintings of this period and later: some areas are very sensitive to solvents like acetone, as if a large percentage of resin had been added to the paint.
Osage Orange (private collection), dated 1934, has extreme problems with very wide traction crackle (fig. 2). In this case, it is less clear that a glossy underlayer caused the crackle, because all layers of the painting are glossy, with many wrinkled areas suggesting that the paint was mixed with excessive amounts of drying oil. Curry's problems with crackle may have been compounded by several additional factors that William McCloy described to us. When McCloy was working with him, Curry often used cobalt drier. McCloy also said that Curry often used a great deal of added medium—a normal palette cup was not large enough to hold his medium, so Curry adapted a flash powder container, which would hold much more, to use as his palette cup.
Detail of John Steuart Curry, Osage Orange, 1934, tempera and oil resin on canvas, 101.5 x 76.3 cm (40 x 30 in.). Private collection
It is interesting that in some of Curry's paintings, it is not only glazes but also opaque body colors that are unusually sensitive to solvents; this is true, for instance, of the 1937 Self-Portrait (Fine Arts Museums of San Francisco). Curry's 1938 sketch for the mural The Homestead (private collection) also has paint that is very sensitive to solvents and is somewhat sensitive to heat. Curry's notebook does not describe the media used for the paint in The Homestead, but in other paintings in the same series he recorded in his notebook that he was using for his underpainting an emulsion made of egg and damar, with no oil. A medium containing one-third to one-half damar was then used to finish some of these paintings (Curry 1932–38, 106–7). The presence of so much damar in both lower and upper layers could help explain why even opaque body colors on paintings like The Homestead react so strongly to heat and to solvents.
Curry's grounds varied a great deal. Notebook references to some paintings from 1932 to 1935 describe them as having been painted on a “chalk ground” (Curry 1932–38, 19). This is most likely the Doerner recipe obtained from Emil Gauss cited in the notebook (Curry 1932–38, 98), which Doerner also called a “half chalk” ground, consisting of chalk, zinc white, glue, and oil (Doerner 1934, 23ff.). In 1936 Curry made, on a large scale, a version of the half-chalk ground containing lead white as well as zinc white for the murals in the Department of Justice in Washington, D.C. (Curry 1932–38, 104). McCloy remembers making half-chalk grounds for Curry in the early 1940s; he said that he had to “beat like hell” to combine the oily and aqueous components, and then it was a horrible job to clean up the tools that had been used to mix them.
The 1938 sketch for The Homestead described above also has an unusual ground. The tacking margins of the fabric show what appears to be an oil tide line from the application of the ground, but the ground is also sensitive to water. An entry in Curry's notebook explains the behavior of the ground on this painting, which is described as two-thirds glue mixed with titanium white and zinc white, and one-third “Dutch Boy” (Curry 1932–38, 106). “Dutch Boy” refers to a lean mixture of lead white and linseed oil manufactured by the National Lead Company. A metal paint bucket found by the authors (not connected with Curry, but from approximately this time period) labeled “Dutch Boy Soft Paste White Lead” lists the ingredients as lead white 88%, linseed oil 10%, and turpentine 2%. Directions on the side of the bucket indicate that the paste was intended to be mixed with 50% to 100% additional linseed oil to make a house paint.
But Curry's notebook shows that by 1938 he was also using commercial products made by Benjamin Moore & Company for the grounds of some of the sketches that he made for murals at the Department of the Interior (Curry 1932–38, 106). There is evidence that this idea came from Grant Wood (1892–1942). Curry's papers contain an undated sheet of Mrs. Grant Wood's stationery that says: “Moore's White Enamel Underbody. Pour off liquid before using as a thick paste” ([Wood] undated). Two letters that chemist H. D. Rasmussen of Benjamin Moore wrote to Curry in January 1938, in response to questions from Curry, verify the connection with Grant Wood. Rasmussen told Curry that Grant Wood used Moore's primer sealer, but he was not sure whether Wood used it over a layer of size or directly on the canvas. Rasmussen suggested that Curry use Moore's underbody enamel with its full amount of liquid, rather than pouring off the liquid,“as you are now doing.” He said that if a heavier body is needed, Curry should consider Moore's Nuwhite. Rasmussen also told Curry that primer sealer was made of mostly tung oil (China wood oil), with the balance linseed oil, and that Moore's underbody enamel was made of the same oils but with 7˝% resin content as well (the resin is not specified) (Rasmussen 1938a, b; see also Horns and Parkin 1995 and Martin 1995). In 1938, Curry recorded that he was using both primer sealer and underbody enamel, sometimes a layer of primer sealer followed by a layer of unadulterated underbody enamel, but often adding lead white pigment to the underbody enamel in amounts ranging from “a little” to 50% (Curry 1932–38, 105–6). In 1938, Curry's notebook indicates that he was also experimenting with “Martz mixture” for grounds, consisting of lead white and titanium white in equal parts, with 12% linseed oil added (Curry 1932–38, 107). Again, the pattern is one of a compulsive experimenter who sought out recipes from a variety of sources and felt free to modify recipes if they did not suit his needs.
2.5 CURRY AND VARNISHING
We found fewer descriptions of Curry's varnishing practice than might have been expected from a man who was so concerned about every other aspect of a painter's technique. Two letters make it clear why this was so, especially with regard to his easel paintings. The letters indicate that Curry left decisions about varnishing easel paintings to his dealers, but they also make it clear that he assumed that his paintings would be varnished once they had dried adequately. A 1938 letter from Maynard Walker of Walker Galleries to Curry reads: “The football pict. doesn't seem to need varnishing very much but we will take care of it. … Reggie Wilson just came in and agrees with me about the football picture. It is dried in only a few little spots which are not visible at all. I think it would be better off if it were not varnished for a while” (Walker 1938). In 1941 Curry wrote to R. Lewenthal, his new dealer at Association of American Artists, about paintings he had sent to New York the previous month: “A great many of these paintings are fairly green and should be given a final varnish about a year from now” (Curry 1941a).
Many of Curry's easel paintings that were in his studio at the time of his death (and remained in his widow's possession) are unvarnished, the exceptions being paintings that were treated at a later time by conservators. The surfaces of the unvarnished paintings vary a great deal; some are quite glossy, others are quite matte, and many have both matte and glossy passages. It seems possible that the survival of so many unvarnished easel paintings by Curry could be partly explained by the trajectory of his career: after a period of success in the 1930s, he had fewer exhibitions and sales during the rest of his life, and it is likely that the unvarnished pictures would have been varnished if and when they had been wanted for exhibition or sale.
Curry's mural paintings are a separate case. There are several references in Curry's papers to his struggle to find an appropriate varnish for murals. He wrote about his 1937–42 murals in Topeka, Kansas:“When I had finished the painting, the coloring was very brilliant and shiny, and I felt it desirable to dim it a little by covering the surface with a special kind of wax varnish” (Curry undated b). But it seems that neither Curry nor John Mathieson, one of his assistants in the project, was entirely happy with the wax varnish they applied, in part because it adversely affected the appearance of the dark blue sky (Curry 1942a; see also Mathieson 1945). About the same time, Curry received a letter from Thomas Hart Benton, apparently in response to a question about wax varnishes, in which Benton included his recipe for a beeswax-carnauba coating (Benton ca. 1943). Curry also used on his murals the new matte varnishes Matvar 39 and Matvar B, which had been developed by F. Weber Company. Curry's papers contain correspondence with F. W. Weber about problems he was having with the application of these varnishes, as well as explanations on Weber's part about how they were intended to be used on mural paintings (and on tempera paintings as well) (Curry 1942b; Weber 1942a; Curry 1945a; Weber 1945).