JAIC 1992, Volume 31, Number 2, Article 1 (pp. 147 to 160)
JAIC online
Journal of the American Institute for Conservation
JAIC 1992, Volume 31, Number 2, Article 1 (pp. 147 to 160)





The four fungal strains were individually streaked onto sterile agar plates at pH 5.6, as described in section 2.2, and incubated at 25C. (A pH of 5.6 was used because this is the pH of the medium when prepared as described.) After 3 days of incubation at constant temperature and exposure to fluorescent light for 8 hours a day, all fungi had grown significantly both on the agar and the paper, covering an area of several square cm. Removal of the fungus by scraping with a scalpel revealed significant staining of the paper by each fungus (fig. 2). From the homogeneous appearance of the fungi as well as their characteristic stains (see below), we ascertained that the individual fungi were not contaminated by other micro-organisms. Thus, growing the fungi in this manner provides a rapid means of obtaining paper stained by a specific fungus.

Fig. 2. Stains produced by Chaetomium globosum, Fusarium oxysporum, Alternaria solani, and Penicillium notatum

Growth on enriched agar medium provides both essential nutrients and water needed for rapid fungal growth. However, this medium also supports the growth of many other micro-organisms that grow faster than the fungi described here. It is therefore critical that the agar and paper have been sterilized and that the fungi are introduced to the agar aseptically.

All four fungal species produce characteristic pigments that stain paper or any medium on which they are grown. Many of these fungal pigments have been isolated, and in most cases they contain a complex mixture of chemicals that strongly absorb visible and ultraviolet light. The stains produced on paper by the four fungi (see fig. 2) are consistent with previous reports of pigments from these fungal species. Penicillium notatum produces a yellow-green stain on paper under our growth conditions consistent with the yellow compounds secreted by this species (Rothe 1950; Wolf and Wolf 1947). The stain apparently penetrates the paper, because its intensity is the same on both the back and front of the paper. We subsequently found that the isolated stain, extracted from the growth medium, is bright yellow (Szczepanowska and Lovett, unpublished). The greenish appearance of the fungi corresponds to the color of the spores, which are embedded on the paper surface and difficult to remove mechanically. Alternaria solani produces a pronounced black stain that adheres very strongly to the paper, consistent with reports of black pigments produced by this fungi (Stoessl 1969; Zaprometova et al. 1971). In fact, it is not possible to mechanically remove the fungus from the paper, and we cannot determine whether the stain is secreted on to the paper or simply a pigment contained within the fungus itself. There is evidence that the pigment produced by A. solani is not secreted by the fungus (Zaprometova et al. 1971). Fusarium oxysporum produces a purple-pink stain that remains on the paper after the fungus is mechanically removed. This stain could be due, in part, to the presence of fusarubin, a red pigment secreted by Fusaria(Ruelius and Gauhe 1950). The stain appears to penetrate the paper, although the intensity on the surface in direct contact with the fungus seems to be greater. Chaetomium globosum produces a yellow stain after about 3 days; the stain becomes grayish-brown in another 2 days. This stain could be due, in part, to the presence of tomichaedin, a yellow pigment extracted from the mycelium of Chaetomium species (Itabashi et al. 1955).

Scanning electron microscopy affords a relatively simple way to examine the effects of fungal growth and stain production on the paper at a microscopic level. Figures 3–7 show scanning electron micrographs of unstained paper and paper samples stained by the different fungi. Fungal debri, including mycelium and spores, are evident on paper stained by a fungal species. Although all four species produce enzymes that digest cellulose (Agrawal et al. 1963; Peterson 1963; Rautela and Cowling 1966), comparison of unstained paper with stained paper suggests that in most cases the destruction of paper fibers is not severe (see figs. 3–6). It is likely, however, that under our growth conditions, the components of the medium are preferred sources of nutrients; this would not be the case when fungi infest paper artworks.

Fig. 3. Scanning electron micrographs of unstained paper. (Photo is 60% of original size.)

Fig. 4. Scanning electron micrographs of paper stained by Alternaria solani. (Photo is 60% of original size.)

Fig. 5. Scanning electron micrograph of paper stained by Fusarium oxysporum (Photo is 60% of original size.)

Fig. 6. Scanning electron micrographs of paper stained by Penicillium notatum. (Photo is 60% of original size.)

Fig. 7. Scanning electron micrograph of paper stains digested by Chaetomium globosum. (Photo is 60% of original size.)

The exception is paper stained by C. globosum; after 8 days of growth on paper, C. globosum has apparently caused extensive digestion of the paper matrix (see fig. 7). After 30 days the paper was completely digested by this highly cellulolytic fungus.


A wide body of evidence indicates that optimum fungal growth occurs in acidic media; however, the range of pH that will permit growth varies with the species and the composition of the culture medium (Wolf and Wolf 1947). The pH dependence of pigment production probably depends on a variety of metabolic effects as well as the acid-base-indicator properties of fungal pigments (i.e., their ability to exhibit color changes at different pH values). An understanding of the pH dependence of fungal growth and stain production could suggest possible methods of paper treatment that would preclude fungal growth and stain production.

The four fungal species were grown on paper equilibrated with medium at pH values ranging from 5 to 8; growth and stain produced to the paper were examined, and the results are summarized in table 1. Although the extents of growth and stain production were not rigorously quantified, the differences observed were sufficiently distinct that trends can be clearly seen through a qualitative assessment. Both growth and stain production were assessed relative to the pH 5.6 results described above and divided into three categories: (1) maximum growth-staining (++) corresponding to levels visually indistinguishable from the pH 5.6 results; (2) moderate growth-staining (+) corresponding to levels that were clearly less than half those observed at pH 5.6; and (3) the absence of visually detectable growth or stain production. There was no measurable change in pH of the medium during the time the fungi were growing. This is not surprising because the enriched medium is expected to contain a high concentration of substances that can act as buffers.


All species except C. globosum showed a strong dependence of stain production on pH, regardless of growth rate; in every case stains were most intense at pH 5. P. notatum exhibited no detectable difference in growth rate over the entire pH range; however, stain production showed a definite pH dependence. Stain production was most intense at pH 5; a bright yellow stain was visible on the verso surface after 4 days of growth. By contrast, at pH 8 there was no color produced, and the fungus itself was white and had a leathery texture. A. solani showed a marked pH dependence for both growth rate and stain production; at pH 5 a very intense black color was visible on the verso surface, whereas no color was apparent on the substrate at higher pH values. At pH 8 there was no detectable growth of A. solani after four days. F. oxysporum showed similar growth rates over the entire pH range. A purple stain was secreted to the paper when it was grown at pH 5 and pH 6, but the stain was more intense at pH 5. Very little stain was secreted to the paper at pH 7 and pH 8; however, at pH 8 the mycelium were more intensely purple than at the lower pH values. C. globosum did not show any pH dependence of either growth rate or stain production; growth and stain production were comparable at all pH values tested.


The temperature dependence of fungal growth has been well documented (Wolf and Wolf 1947). Studies of a wide variety of fungal species indicate that for each fungus there exists a minimum, a maximum, and an optimum growth temperature, the so-called cardinal temperatures. The first two of these correspond to temperatures that inhibit fungal growth, while the latter is the temperature that presumably encourages maximum metabolic activity. Most of the fungal species that have been studied have minimum temperatures between 3C and 5C, maximum temperatures between 30C and 37C, and optimum temperatures between 22C and 29C.

Based on these reported temperature ranges, we studied the effect of temperature on growth as well as stain production for each of the four fungal species (table 2). As described above, a qualitative assessment of the differences in growth rate and stain production was deemed sufficient. We chose three temperatures representative of the three cardinal temperatures: 4C, 25C, and 37C. After several months there was no evidence of growth for any of the strains at 4C. Surprisingly, C. globosum grew rapidly at 37C, and stains were apparent after the second day; in fact, growth and stain production by this fungus were greatest at 37C. In contrast, this temperature completely inhibited the growth of P. notatum and markedly reduced the growth rates of F. oxysporum and A. solani. The optimum temperature for both growth and stain production by these three fungi was 25C. At 37C F. oxysporum grew slowly on culture media as well as on paper but only produced stain to the medium, not on the paper. On the other hand, A. solani did not grow on paper at 37C but grew slowly and produced stain on culture media.


Our results for F. oxysporum are consistent with previous studies (Hulme and Stranks 1976; Linfield 1986). However, our results disagree with those of Hulme and Stranks (1976) for C. globosum, for which they reported an optimal temperature of 25C, this disagreement may be due to differences in growth medium and light conditions.

Overall, the results are not surprising with the exception of C. globosum. Unlike the pH dependence results described above, both growth rate and stain production showed essentially the same temperature dependence.


The effects of radiation on fungal growth are widely varied and depend on both the wavelength of radiation and the fungal species (Wolf and Wolf 1947). For example, in some cases UV and visible light have stimulatory effects on fungal growth, while in other cases the effects are inhibitory. The results reported here deal only with varying exposure to fluorescent light (0, 8, or 16 hours per day) at 25C. For all four fungal species growth was better with exposure to light, although there was no detectable difference in growth when light exposure was increased from 8 hours per day to 16 hours per day. There is no apparent effect of light exposure on stain production by A. solani, C. globosum, or P. notatum; however, stain production by F. oxysporum showed a marked dependence on light. Figures 8 and 9 show F. oxysporum grown for 5 days either in the dark or exposed to light for 16 hours per day. When grown in the absence of light, there is a reduction in the amount of pinkish orange pigment present in the mycelium that is obvious from the top view (see fig. 8). Moreover, growing the fungus in the dark results in a significant decrease in the amount of pigment secreted to the medium, as seen by viewing the bottom of the culture plate (see fig. 9).

Fig. 8. Growth of Fusarium oxysporum in the absence of light (left) and as exposed to light for 16 hours per day (right) as viewed from the top of culture plate

Fig. 9. Growth of Fusarium oxysporum in the absence of light (left) and as exposed to light for 16 hours per day (right) as viewed from the bottom of culture plate


Several components of the pigments from the four fungal species have been isolated and characterized. Xanthocillin and Citronen are yellow-colored antibiotics produced by P. notatum, which are probably responsible for some of staining by this fungus (Rothe 1950; Wolf and Wolf 1947). (Xanthocillin is actually a complex that consists of at least three distinct components, xanthocillins X, Y1, and Y2). Chaetomidin, a pigment extracted from the mycelium of Chaetomium species, was found to be identical to the bright yellow dye, oosprein, originally isolated from Oospora colorans(Itabashi et al. 1955). Fusarium species produce the red pigment, fusarubin (Ruelius and Gauhe 1950). Intracellular black pigments, which are not secreted under laboratory conditions, have been isolated from Alternaria species (Stoessl 1969; Zaprometova et al. 1971). Preliminary characterizations of these pigments suggest a resemblance to humic acids and melanins (Stoessl 1969). Pigments isolated from the mycelium of A. solani were shown to contain a mixture of anthraquinones and have been named altersolanols A and B (Zaprometova et al. 1971).

Based on the general solubilities of these and other dyes, we tested the stain-removing abilities of the following solvents on stains produced by each of the four fungal species: dimethyl sulfoxide, 1,4-dioxane, N, N-dimethylformamide, triethylamine, pyridine, and 1,4-butanediol diglycidyl ether. Short-term exposure of any of these solvents to the stained paper did not have any detectable effect on any of the stains. However, after 24 hours of soaking, 1,4-dioxane completely removed the stain produced by F. oxysporum and significantly reduced stains produced by P. notatum and C. globosum. Twenty-four hour exposure to N,N-dimethylformamide significantly reduced the stains produced by C. globosum and P. notatum and slightly reduced the stain produced by F. oxysporum. There was also a slight effect of this solvent on the surface of the stain produced by A. solani. Treatment with pyridine for 24 hours had a slight effect on the stains produced by C. globosum and F. oxysporum. Dimethylsulfoxide, triethylamine, and 1,4-butanediol diglycidyl ether had no detectable effect on any of the stains. Attempts at stain removal using some more common solvents such as acetone, ethanol, and xylene also had no detectable effect on stained paper.

Following exposure to the stained paper, the solvents were analyzed for the presence of pigment by recording their UV-visible spectrum (180–800 nm) against a pure solvent blank. In every case where stain had been reduced or eliminated, the solvent had absorption peaks that were not present in the pure solvent.

To assess the effects of the solvents on the characteristcs of the paper, samples of paper were soaked in 1,4-dioxane, N,N-dimethylformamide, or pryidine for different lengths of time and then tested by Crane and Co., Inc. Figure 10 shows the results of dry and wet tensile tests (tests of paper strength) on samples that had been exposed to solvents for the indicated time periods. Relative to the untreated paper (at time = 0), there does not appear to be any significant change in tensile strength after continued exposure to any of the solvents. To assess paper sizing, an ink test was conducted by Crane and Co. in which the ratings are given as good, fair, or poor: results with untreated paper, paper treated with N,N-dimethylformamide (1, 3, 6, or 18 hours), and paper treated with 1,4-dioxane (1, 3, or 6 hours) were rated good; results from short-term exposure to pyridine (1 hour) or 18-hour exposure to 1,4-dioxane were rated fair; and results from long-term exposure to pyridine (3, 6, or 18 hours) were rated poor. Thus it appears that N,N-dimethylformamide and short-term exposure to 1,4-dioxane do not significantly affect paper sizing, whereas exposure to pyridine, even for short periods of time, is detrimental to sizing. Other test results showed a great deal of variability that was independent of solvent exposure time and are therefore not included here.

Fig. 10. Effects of solvent exposure on tensile strength of paper. Results from dry tensile tests (open symbols) and wet tensile tests (closed symbols) after soaking in 1,4-dioxane (▪,□), N,N-dimethylformamide (♦, ⋄), or pyridine (▴, ▵) for 0,3,6, or 18 hours

It should be noted that the solvents that were effective at removing fungal stains present varying degees of toxicity and should be handled with extreme care: 1,4-dioxane is a flammable liquid that produces a harmful vapor and may be irritating to skin, lungs, and mucous membranes; N,N-dimethylformamide produces a harmful vapor and is irritating to skin, eyes, and mucous membranes; pyridine is a flammable liquid that produces a harmful vapor and is irritating to skin and the respiratory tract. All of these solvents should be handled with gloves and used in a ventilation hood.

Copyright 1992 American Institute for Conservation of Historic and Artistic Works