6 TEST CANVASES

The most interesting manifestation of Curry's spirit of skeptical inquiry is a series of test canvases, the majority of which are now in the Archives of American Art and one of which is at the Worcester Art Museum (fig. 6). F. W. Weber suggested in 1942 that Curry paint out samples of the Maroger media (Weber 1942b), but the evidence of notations on the canvases themselves reveals that Curry had already begun his testing program by that time. In November 1939, Curry made tests in which he added damar and sun-thickened linseed oil to tube oil paint, and during 1941 he painted out samples of many different white oil paints, with and without various additives (including Maroger media). The brands of white paint tested included Weber, Bocour, Schmincke, Blockx, Dutch Boy, and Sherwin-Williams, as well as lead carbonate from “Merk Drug,” which Curry mixed with linseed oil. Curry also carried out experiments with egg, egg-oil mixtures, egg-oil–gum arabic mixtures, oil-damar-water mixtures, Venice turpentine, casein paints, and oil-casein mixtures. In studying the test canvases, one gets the impression that Curry was making up his own variations rather than directly copying the recipes of others. Not all of the samples are dated, but new tests continued to be added to the canvases until at least 1944.

Fig. 6.

Detail of test panel painted by John Steuart Curry, 1939–1944, oil (with other added media) on canvas. Archives of American Art, John Steuart Curry Papers

Tubes of paint that were owned by Curry also survive (at the Worcester Art Museum, the gift of William McCloy). The majority of these are products of Weber, Bocour, Schmincke, and Blockx, but there are also tubes carrying the labels of Rembrandt (made by Talens), Winsor & Newton, Grumbacher, Shiva, and Chemically Perfect Pigments (made by Rich Art Color Company, New York).

Interpreting the results of Curry's test canvases is somewhat difficult because Curry did not always write down the details of the recipes he was testing. For instance, lead carbonate from Merck plus linseed oil was the most discolored of the white samples, but Curry does not tell us the ratio of oil to pigment or the type of linseed oil used.

However, some conclusions can be drawn from the test canvases. As might have been expected, adding medium of any kind to various lead white oil paints made them discolor more than paint straight from the tube, and adding more medium made the samples discolor still more. Black oil was not noticeably worse than other additives in producing discoloration, although it is difficult to make precise comparisons because Curry did not state the amounts of each additive. Again, as might have been expected, zinc white and titanium white have generally discolored less than lead white. Weber Permalba stands out as having remained noticeably whiter than any of the other white samples. Samples of Permalba having gum arabic and damar medium added to them have remained quite white, while samples of lead white with the same additives are noticeably dark. In 1941, Curry experimented with adding small amounts of black oil or cobalt drier to Permalba white, without any obvious ill effects on the white color of the samples.

Curry's experiments prompted an interesting exchange of letters with Leonard Bogdanoff of Bocour Inc. in 1941. Curry explained why he had requested that his name be removed from a list of artists recommending Bocour colors:

You need not feel personally hurt about my not using your colors. If you remember, I complained several times about the dirty oil in your white. You said that it made no difference.

I found the paintings made with your color beginning to get very dingy and spotty. I have made some very complete tests of various lead whites and found yours to be among the poorest. I am sorry to tell you this, but you will remember my complaints. I found your other colors satisfactory. (Bogdanoff 1941; Curry 1941c).

In fact, one sample of Bocour Cremnitz white that Curry painted out in 1941 is darker than many of the other whites, although another sample similarly labeled (but not dated and perhaps applied at a different time) is less dark.

Other interesting observations about Curry's test panels are that an “egg wash” is very noticeably dark and brownish, and a complex test area containing lead white, gum arabic, and zinc white is noticeably browner in the area labeled “same with turpentine.” Two samples of Permalba that had gum arabic added to them are disfigured by a pattern of fine mechanical crackle, confirming what Weber told Curry about the brittleness and lack of adhesion of gum arabic, although there is no way of knowing whether Curry saw this effect during his lifetime or if it influenced him against the Maroger recipes that contained gum arabic.