13. FOXING 113.1 Purpose 113.2 Factors to Consider 113.2.1 Causes 1A. Fungal Activity 2B. Metal-Induced Degradation 7C. Condensation. 9D. Multiple Causes 1013.2.2 Origin/Occurrence. 11A. Occurrence Related to Manufacture of Paper 11B. Causes Related to Storage 12C. Causes Related to Dampness 1213.2.3 Classification of Foxing 12A. "Bullseye" 13B. "Snowflake" 13C. Stains Confused with Foxing: Offset or Migration 1313.2.4 Methods of Identification 14A. Visual Examination in Normal Light 14B. Transmitted Light 14C. Microscopic Examination 14D. Ultra-violet Fluorescence 14E. Spot Tests 15F. Analytical Instrumentation 1513.2.5 Effects of Foxing 17A. Support 17B. Media 18

13.3 Materials and Equipment 1813.3.1 Examination Equipment 18A. Near Ultra-violet ("Black Light") 18B. Spot Test Reagents 18C. Cultures 18D. Stereomicroscope. 1813.3.2 Treatment Equipment 18A. Chemical Reagents 18B. Tools 18

13.4 Treatment Variations 1913.4.1 Environment 19A. Housing 19B. Relative Humidity (RH) 19C. Temperature. 19D. Ventilation. 1913.4.2 No Intervention 1913.4.3 Fungicides 2013.4.4 Alkaline Washes 20A. Neutralization of Acidity 20B. Reduction of Staining 20C. Treatment of Metal Ions 21D. Pre-treatment for Bleaching 2113.4.5 Metal Removal or Inactivation 21A. Mechanical 22

B. Chemical 22

13.4.6 Bleaching 23A. Chlorine Dioxide 23B. Hydrogen Peroxide 24C. Calcium Hypochlorite 24D. Chloramine-T 25E. Sodium Borohydride 25F. Natural or Artificial Light 2513.4.7 Conservator's Observations of Foxing 25A. Early Papers and Artists 26B. 17th and 18th Century Papers and Artists 26C. 19th Century Papers and Artists 26

D. 20th Century Papers and Artists 26E. Printing Papers 27

F. Chine Coll Papers 27G. Japanese Papers 27H. Effects of Light 2813.4.8 Observations on Current Conservation Practices 28A. Examination. 28B. Treatment 29

13.5 Bibliography 3313.6 Special Considerations: Reverse Foxing 37

13.6.1 Definition. 3713.6.2 Description. 3713.6.3 Treatment Warnings 38

13. Foxing, page 1

13. FOXINGThe term foxing is derived from the rusty red color of Reynardthe fox and its use was first noted in 1848 [Meynell and Newsam1979, 567; Carter 1980, 105]. Foxing is a descriptive term forscattered spots commonly reddish-brown in color, but also appliedto spots of other coloration ranging from yellow to black. Itshould be distinguished from visible surface colonies of moldgrowth, which may result in paper stains of a wide range ofcolors, though both may be present concurrently. In Japan,foxing is known as hoshi, which literally means "stars"[Engelbrecht 1991, 62].It may be easier to define what foxing stains (as identified bymost conservators) are not. They are not the mold stains, withor without surface growth, which severely deteriorate the paperand cause a variety of colorations. They are not offset stainsfrom contact with another paper or printing ink. They are nottide lines of liquid stains. They are not acid stains migratingfrom secondary material, although poor quality secondarymaterials may accelerate the foxing stain phenomenon.

13.1 Purpose1. To provide criteria for differenting foxing problems from

other stains in paper.2. To ascertain, to the extent possible, the cause(s) of a foxing

stain in paper.3. To record established methods of identification used by both

conservators and research scientists.4. To identify areas needing further research.5. To assemble treatments appropriate to various foxing stains.

13.2 Factors to Consider13.2.1 Causes

Despite investigations spanning almost sixty years thereremains confusion and uncertainty as to what causesfoxing, whether there is a single cause or multipleones, and whether there is more than one type of foxing.There are currently three major explanations for foxingwhich have been proposed most often: a) fungal activity,b) metal-induced degradation, and c) multiple causes.Recently, a fourth explanation has been proposed whichattempts to explain foxing stains within the context ofgeneral discoloration of paper caused by the interactionof moisture and cellulose.Unfortunately, most of the published research on foxingneglects to provide accurate and complete information

13. Foxing, page 2

concerning size, shape, depth (within the sheet), orfluorescence pattern (or absence thereof) whendescribing foxing. In the future, it is hoped thatinvestigators will use Cain's proposed classificationsystem (see 13.2.3), making it easier to integratedisparate research. Although researchers claim theirstains had the "typical coloration characteristic offoxing," this is fairly meaningless, as there is no wayof judging the uniformity of this coloration. It ispossible that researchers are finding different causesfor foxing because the spots tested are different, eventhough they are all termed "foxing". Hey has evensuggested removing metal induced foxing from the foxingcategory altogether as this is "metal induceddegradation". Additionally, researchers proposing onecause over another often neglect to investigate bothcauses equally, lending an unstated bias to theirfindings of which the reader must be aware. Followingis a summary of investigations and theories culled fromconservation literature.A. Fungal Activity

There are approximately 100 fungoid speciesdesignated as 'paper-attacking' though somespecialists suggest that many more of the more than30,000 species of fungi would attack paper [Gallo1963, 57; RK]. Not all of these species arenecessarily associated with foxing.Some of the micro-organisms habitually associatedwith paper are found in the raw materials used toform the paper. These micro-organisms can remainlatent for months or years awaiting the appropriateconditions for growth. Another likely means ofinfection of paper is through air-borne spores.Certain fungi have been associated not just withpaper generally but with foxing spots specifically.From old books and manuscripts showing thecharacteristic foxed discoloration, Beckwith et al.isolated fifty-five different fungi [Beckwith et al.1940, 301]. These they sampled directly from foxedareas of the paper. From foxed spots Arai foundtwenty-five strains located specifically within thefoxed areas of paper and, in fact, could not findevidence of the fungi in any unfoxed portions of thepaper [Arai 1987, 1165]. Although these strains wereall found in foxed spots, only seven were found tocreate browning stains in Arai's experiments.

13. Foxing, page 3

1. ExplanationsThe following explanations for formationmechanisms of foxing spots by fungi have beenproposed.a. Color pigments are secreted by the mycelia, the

vegetative branches of mold. These pigmentshave various chemical compositions but aremainly composed of carotenoids andanthraquinones [Gallo 1963, 58]. Beckwith etal. state that pigments may be only a minorcause of foxing because the chromogenic tintsproduced by various species of mold are notcharacteristic of foxing [Beckwith et al. 1940,306]. Nol et al. believe fungi producepigments and that coloration can be intensifiedwithin a foxed spot by certain combinations offungi species which alone may not producestrong coloration [Nol et al. 1983, 22]. Thisexplanation does not show a clear relationshipbetween foxing and iron.

b. According to one source, micro-organisms"develop at the expense of the glue materialsforming hygroscopic areas on the paper in whichthe water soluble degradation products of thecellulose accumulate. They assume a red-browncolor in damp surroundings" [Ambler and Finney1957, 1141 as quoted in Gallo 1963, 26]. Thistheory is also consistent with Meynell andNewsam who found "foxed areas invariably showedfungal hyphae weaving around but not withinindividual cellulose fibers. The cellulosefibers within foxed lesions appeared normal.The lesions, however, wetted more easily thanthe rest of the paper and stainedinstantaneously with Coomassie blue, indicatingthe sizing was participating in the fungalgrowth [Meynell and Newsam 1979, 567]. Thisexplanation is inconsistent with findings byCain and Nol et al. who found evidence ofcellulolytic activity and had to assumecellulose degradation was occurring [Cain 1983;Nol et al. 1983, 23]. This explanation showsno clear relationship between foxing and iron.

c. Specific fungi tested to date have been foundto produce foxing in laboratory experiments.They do this by secreting malic and otherorganic acids as well as amino acids. Theseacids are deposited on, and then attack thecellulose and/or sizing. This produces cello-oligosaccharides and glucose which incombination with amino acids produce a browning

13. Foxing, page 4

reaction known as the Maillard reaction [Araiet al. 1988, 12]. Each ingredient alone willnot induce a browning reaction, but they willin combination. This explanation does notdemonstrate a clear relationship between foxingand iron.

d. Foxing is the visible sign, by production ofcolor, of deterioration within paper. Althoughthe breakdown of paper may result from manycauses, fungi constitute a most importantmenace to the preservation of paper. Growth offungi in paper and development of color arefurthered by iron and by certain sizings andfillers [Beckwith et al. 1940, 305]. Thisexplanation acknowledges foxing may havemultiple causes requiring some iron content forfungi to create sufficiently strong coloration.

e. The fungi produces an enzyme or ferment whichbrings about a chemical change. Thisenzyme/ferment may be diffusible when moistureis present, explaining effects visible somedistance from the site of the formation of theenzyme. The enzyme, in turn, may bring aboutits chemical change and the products of itsactivity may also diffuse some distance,causing local discoloration [Iiams and Beckwith1935, 415].It is generally known that fungal growth breaksdown its support by releasing enzymes. It maybe that the circumstances that cause foxingalso include some form of enzyme attack ofcellulose. "Once growing, fungi cause damagein several ways. Actual damage to material iscaused by the release of enzymes outside theorganism (extracellular enzymes). Theseenzymes break down long-chain molecules such ascellulose or proteins into chemical unitssufficiently small to be absorbed through thecell membrane into the cell. The action ofextracellular enzymes is independent of thefungus; they can act even if the organism iskilled or removed" [Allsop 1985, 532]. Thisexplanation shows no relationship betweenfoxing and iron.

2. TypesThe following fungi have been culturedspecifically from foxing spots and create newbrowning stains when reinoculated into paper underlaboratory conditions.

13. Foxing, page 5

a. Beckwith et al. [1940, 301] list genus1. Alternaria2. Aspergillus3. Byssochlamys4. Chaetomium5. Fusarium6. Hormodendrum7 Monilia8. Mucor9. Penicillium10. Stemphylium

b. Arai [1987, 1166] lists species1. Aspergillus penicilloides2. Eurotium herbariorum

c. Nol et al. [1983, 22] lists species1. Aspergillus carneus2. Aspergillus flavus Link (with Sclerotia)3. Aspergillus flavus Link (lacking Sclerotia)4. Aspergillus fumigatus Thom5. Aspergillus niger van Tieghem6. Aspergillus terreus var. aureus Thom &

Raper7. Aspergillus tamarii8. Gliocladium roseum (Link) Thom9. Penicillium funiculosum Thom

d. Cain et al. [1987, 24] found speciesAspergillus repens

3. Factors favoring growth and/or color formationa. Relative Humidity (RH)

There is no minimum degree of RH for growth ofall fungi as the level varies not only with thegenus but with the species as well [Beckwith etal. 1940, 311-313]. The more hygroscopic thematerial (i.e. paper vs. leather) the lower theroom RH can be in order to permit microbialgrowth [Gallo 1963, 59]. It is important toremember that a micro-climate, vastly differentfrom the overall room RH, can exist within thepaper's structure. Unsized tissues, forexample, interleaved into books may absorb andretain water over a long period of timecreating localized high humidity relative toatmosphere; this could enable fungus to developeven when surrounding RH is below 75%. Fungalhyphae may transport water from higher to lowerareas of RH, perhaps centimeters away [RK].

13. Foxing, page 6

Most research directed at microbial growth onpaper shows that below 70-75% RH the chances ofgrowth for many "paper-attacking species" offungi is low. Arai found that below 75% RH,germination of mold spores of the type isolatedfrom foxed spots is unlikely to occur.Interestingly Arai also found that 84% relativehumidity induced growth better than 94% RH,though this is specific to the particular fungihe isolated [Arai et al. 1990, 805]. Beckwithet al. also found that below 75% they could notproduce any fungal growth with the particularspecies they cultured [Beckwith et al.1940,313]. In contrast to this, Nol et al. foundthree strains, previously isolated from foxedspots, which grew at 55% to 93.5-96% RH. Twoof these strains also grew at 32.5% RH.However, foxing or coloration occurred onlywith one strain under RH conditions rangingfrom 32.5% to 96% [Nol et al. 1983, 24].

b. TemperatureEach species has its optimum temperature forgrowth. Generally, it has been found thatgrowth increases with increasing temperatureand decreases with decreasing temperature.Excessive heat kills most fungi and steam is astandard means of sterilizing cultures in labprocedures.

c. pHAll research has shown that foxing stains aremore acidic than the surrounding paper [see,for example, Arai 1980; Hey 1983; Iiams andBeckwith 1935] .Arai's research suggests that the presence ofamino acids is necessary when inducing foxingand that increasing the concentration the ofamino acids results in darker brown spots [Araiet al. 1990].

d. NutrientsFungi may find nutrients in one or more of thefollowing.1. Cellulose

While some researchers insist thatcellulose is not damaged in foxed areas[Meynell and Newsam 1979, 567], others haveshown conclusively that fungi digest thecellulose [Cain 1983, 16 and Nol et al.1983, 23].

13. Foxing, page 7

2. Sizings and AdhesivesBecause they saw no damage to fibers,Meynell and Newsam claim foxing feeds ongelatin size, not on the cellulose [Meynelland Newsam 1978, 467]. However,observations have also been made that fungiprefer more hygroscopic, unsized papers tothose that are sized [Meynell and Newsam1978, 468; Gallo 1963, 58].Investigation into the influence of fillersand sizes on fungal growth and itsproduction of acids found the following:gelatin, starch and dextrins promotedgrowth and color production [Beckwith etal. 1940, 3307]. There was less acidproduction by fungi feeding on casein androsin than with starch or cellulose alone.

3. OilsEither from the medium of printing ink in atext or that transferred to paper byreaders' or handlers' hands [Meynell andNewsam 1978, 467].

4. Micro-dustLight IntensityGenerally, the growth rates for most fungi arenot sensitive to light intensity. However, nostudy has been made of the relationship offoxing stains to light."Examination of a 1896 thirty-four volume setof Balzac's works on laid cotton paper found'snowflake' fungal foxing in circulated volumesno different from that present in 1896uncirculated, unopened volumes. Previouslyuncut pages were slit in the dark and examinedin the first light exposure in nearly ninetyyears. Apparently dark storage produced thesame pattern, color, and frequency of foxing asoccasional exposure to light [Cain, Stanley andRoberts 1987, 24].

B. Metal-Induced Degradation"Cellulose is directly oxidized catalytically in thepresence of iron, copper, and cobalt compounds, andthe reaction is most rapid at high humidities" [Tang1978, 19]. Metal impurities in paper, specificallyiron and copper, are believed to result fromparticles abraded from the metal equipment and/orfrom contaminated water used in the papermaking

13. Foxing, page 8

process. Additionally, all wood-pulp paper may beexpected to contain iron, as it is naturally presentin wood [Beckwith et al. 1940, 302]."In 'bullseye' copper- or iron-induced foxing therole of these two metals is probably that ofoxidative catalyst. Both metals can undergoreversible oxidation-reduction. For example, theyare both found playing such a role in metabolicbiochemical reactions. Iron can alternately beoxidized from the +2 (ferrous) state to the +3(ferric) state and then be reduced back to the +2state as it plays the role of oxidizer. Copper cando the same between the +1 and +2 states. Thin-layerchromatographic studies show the extracts of'bullseye' foxed and unfoxed paper to have all ormost of the same bands. This further suggests ironand copper act to catalyze (accelerate) the oxidativedegradation of paper" [Cain 1983, 15; Cain andKalasinski 1987, 57]. In a tally of metal-inducedfoxing, analysis showed that twenty-seven wereinduced by copper and copper alloys to over 200induced by iron [Cain and Miller 1982, 7].1. Iron

a. Coloration"The very color of foxing connotes the presenceof iron" [Iiams and Beckwith 1935, 412]. Ironions create yellow-brown spots and Tang foundthat "there is a trend for darkness of thefoxing spot to increase with increasing ironcontent; the highest concentration of iron wasnoted in the center of the spots, with themetal concentration decreasing. . . as thedistance increased from the center" [Tang 1978,24, 26].

b. OccurrenceIt would be very difficult to find any paperwithout some degree of iron [MH]. Numerousresearchers have identified iron ions withinfoxing stains and found a significantly greaterconcentration of iron in the foxed areascompared to surrounding paper [Cain 1983, Cain1988, Cain and Miller 1982, Cain and Miller1984, Daniels 1988, Gallo and Hey 1988, Tang1978, Tang and Troyer 1981]. One study,however, found no difference between foxed andunfoxed areas [Press 1976, 29]. This wascorroborated by Tang, who found that in somefoxed papers there was no difference in iron(or other metal ion) concentration [Tang 1978,28]. While concentrations greater than 500 ppm

13. Foxing, page 9

have been identified with undesirable spots,Hey suggests that "if iron is involved it isnot its total concentration that is importantbut rather its availability to participate inreactions or its effective solubility" [Tang1978, 28; Hey 1983, 341].

c. FormResearch indicates that iron in paper is foundentirely in the ferric, rather than ferrous,form [Beckwith et al. 1940, 303].

d. ActivationIron will not corrode below 70% RH, but in thepresence of ions such as chloride, storageneeds to be at 40% RH or lower to avoidcorrosion. Hey suggests that "there is astrong chemical possibility that heavy metalspresent in the paper in a quiescent state willbe activated by washing with an acid water,when this is not followed by deacidification"[Hey 1979, 68].

2. CopperDaniels and Meeks describe copper-related foxingas varying in size "from small spots with noapparent nucleus and only a brown diffusediscolouration, to large spots of about 5 mmdiameter with black dendritic patterns or greencorrosion products; these spots include an outerring of brown discoloured paper" [Daniels andMeeks nd., 2]. Analysis by EDX revealed that thefoxed areas contained copper, zinc, sulfur, andchlorine, while the unfoxed areas "did not havedetectable amounts of these elements" [Daniels andMeeks nd., 5; see 13.2.4.E.2]. It was concludedthat chloride ions, from original or subsequentbleaching residues, accelerated the corrosion ofbrass (a copper/zinc alloy) inclusions in thepaper. The soluble copper compound was then ableto react with hydrogen sulfide generated in thepaper or absorbed from the atmosphere. The stainwas due to a combination of black copper sulfideand brown copper catalyzed degraded cellulose[Daniels and Meeks nd., 8]. Tang linked copperconcentrations greater than 50 ppm with formationof undesirable spots [Tang 1978, 28].

C. CondensationA modification of the cellulose, often visuallyevident by browning, which takes place at theinterface between wet and dry parts of fibrousmaterials and which is not the result of degradationproducts being carried and deposited by a spreading

13. Foxing, page 10

liquid. "Experiments suggest that the interaction ofair, water and cellulose is responsible for theformation of browning" [Hutchins 1983, 58]. Thisinteraction could occur at sites of temporarymoisture accumulation in the paper. "Depending onthe moisture content of a book, [for example,] itwould be possible for uniform discolouration ofzones, as well as smaller or larger stains, todevelop. All the possible factors that influencecondensation and evaporation would play a role inthis: humidity, temperature, air pressure, paperporosity, and any irregularities in the paper whichcould include folds, tears, and dirt particles; eventhe presence of concentrations of iron or funguscould likewise induce condensation" [Ligterink et al.1991, 51].The above authors speculate on the relationshipbetween foxing and other forms of discoloration (textblock areas, leaf margins). The link is based onobservations of both types of staining (foxing andzonal) appearing together on the same page in manybooks.The condensation explanation for browning is a broadview ascribing moisture and cellulose and possiblyoxygen as the only necessary ingredients to achievestaining. The presence of fungi and/or metals wouldact only as attractive sites for moisture andconsequent browning.

D. Multiple CausesGiven the ubiquitous nature of both iron and fungi inpaper it is quite possible they often act in tandem.Research appears divided (fungal infection vs. metal-induced degradation), and one must keep in mind, whenreviewing each study, whether the presence of a dualcause was fully investigated. Often researchers didnot adequately test for iron when they found fungiand vice versa.A good example of this is the use of the SEM. WhereCain and Miller did not, in one study, find an ironcore using SEM and EDX, they successfully located itusing narrow beam x-ray fluorescence [see 13.2.4.F.1-3]. Other research, however, has relied on SEM aloneto determine that there was no iron (or other metals)present without using other methods to check theirresults.As early as 1935, Iiams and Beckwith proposed a dualcause of spot formation: organic acids secreted bythe metabolizing fungi react with iron present (evenin trace amounts) in the paper to form unstable

13. Foxing, page 11

organic iron salts (organo-ferro compounds) whichdecompose to form iron oxides and hydroxides i.e.brown/rust coloration [Iiams and Beckwith 1935, 414].Iiams and Beckwith also found that adding a 1:1,000solution of iron caused fungal growth which "greatlyexceeded any that had been produced in the laboratorywithout the presence of iron in the culture papers"[Iiams and Beckwith 1935, 414]. Their later researchconfirmed this as well as showing that iron increasesthe degree and intensity of the discoloration whichaccompanies fungal proliferation [Beckwith et al.1940, 303-306]. The resulting brown tint had thecolor of ferric oxide. The presence of casein,gelatin, and starch add to the discoloring effects ofiron.Hey concurred with Iiams and Beckwith and proposedthese dual mechanisms:1. damp -> mold acid -> activation of iron ->

increased acid -> mold death2. damp -> activation of iron -> increased acidity ->

local encouragement of mold -> increased acidity-> death of mold [Hey 1983, 341]

These models suggest that one reason why foxingstains do not cover an entire page might be that theacids secreted by the fungi collect, eventuallyreducing the pH enough to curtail further fungalgrowth.Cain and Miller found that "snowflake" foxingcontained a higher iron concentration than thesurrounding paper [Cain and Miller 1982, 61]. Alater study found hyphae and occasional fruitingbodies in all snowflake fungal foxed areas examined[Cain, Stanley and Roberts 1987, 24]. This suggestsa dual cause.Fungi use iron and copper as co-enzymes. This meansthat they are essential elements. After use, theexcess may be secreted (perhaps as an altered oractivated ion) [RK].

13.2.2 Origin/OccurrenceA. Related to the Manufacture of Paper

The extent of foxing appears to be in directproportion to methods used in the manufacture ofpaper [Iiams and Beckwith 1935, 413]. It is possiblethat the potential for foxing is created when thesheet is first made - the foxing only becomes visible

13. Foxing, page 12

later when storage conditions encourage it. Factorsinclude the poor preparation of fibers, impurities inthe pulp and the water added to it, and poorbleaching with chlorine. On the other hand, papersmanufactured with a high magnesium or calciumcarbonate content are less likely to be foxed. Ithas also been noted that woolen or rayon felts aredamaged continually by microbial attack as shown bySharpely and King. Two species of fungi located indamaged felt fibers are Aspergillus niger andAspergillus furnigatus, both associated with foxingand noted for their cellulolytic capabilities [EM].

B. Causes Related to StorageLigterink et al. proved that foxing stains found inone particular book arose during storage of the loosesheets prior to binding. They noted that the stainpatterns which were not repeated on an adjacent pagewere sometimes repeated on another page later in thebook. By reconstructing the original unfolded, uncutquires it was discovered "that the stain patterns ofsuccessive quires matched up if the unfolded sheetswere laid on top of each other, and could often befollowed down through many sheets in the stack. Thestorage of these unfolded sheets obviously determinedthe form of the stains observed which must havetherefore arisen before binding" [Literink et al.1991, 49]. Interestingly, the stains were probablynot visible at the time of binding as thediscolorations are so great in some sheets they wouldprobably have been discarded by the binder [see also13.4.7.E]. By the same reasoning, book papers whichshow the same foxing pattern through several adjacentpages indicate the foxing began and became visibleafter the pages were bound.

C. Causes Related to DampnessResearch indicates that the internal moisture contentof the paper must be at least 10% for fungal growthto occur [Allsop 1985, 59]. At 80% RH, paper ingeneral absorbs 9-14% water, with more hygroscopicpaper, a lower RH will permit mold growth. Ironalone will not corrode below 70% RH but in thepresence of ions such as chloride, papers must bestored at 40% RH or lower to avoid iron corrosion.

13.2.3 Classification of FoxingIn an attempt to describe the various stains called"foxing" Cain and Miller have developed a classificationof types by shape, color and UV fluorescence examination[Cain and Miller 1982, 1984]. Cain and Miller foundthat foxing was a three-dimensional phenomenon. Intheir investigation of foxing in books, Ligterink et al.found that "All foxing stains are part of three-

13. Foxing, page 13

dimensional stain-structures, which are generated instacks of paper either during the production process orin the book itself...many series of stains begin on onepage and end several pages later, forming sharplydelineated, or sometimes rounded, spatial entities"[Ligterink et al. 1991, 50].A. "Bullseye"

These spots are small and round, with a dark centerand concentric rings. They are red/brown to yellowin color with rings of a paler color. Bullseyefoxing can be further subdivided by UV examination.1. Centers do not fluoresce (appear dark blue/black),

rings fluoresce yellow/orange and pale yellow.2. Centers and rings do not fluoresce (appear dark

blue/black).This type of foxing always has metal cores, which donot fluoresce and appear dark blue/black.

B. "Snowflake"These are spots with scalloped edges and/or irregularshapes which can measure inches across. They arered/brown to yellow in color but sometimes are notvisible in normal light. It is theorized that theadvanced stage of foxing causes coloration while theyounger or dormant stage may not be visible in normallight [Press 1976, 29]. This type of foxing gets itsname from its white fluorescence and snowflake-likeappearance under UV. Snowflake foxing apparently hasa higher iron concentration than the surroundingpaper, but concentration may vary within areas of thefoxing.While only one iron particle of twelve found in"bullseye" foxing showed any presence of fungalhyphae, extensive examination of Balzac volumesshowed no evidence of fungus, either hyphae orsporangia, in unfoxed areas, but hyphae andoccasional fruiting bodies in all "snowflake" foxingexamined [Cain, Stanley and Roberts 1987, 24]. Thesespots appear to occur in association with fungalactivity.

C. Stains Confused with Foxing: Offset or MigrationThe following discoloration are seen by conservatorsto be staining rather than foxing. Thesediscolorations are often found in books andcorrespond in shape to a print or images within aprint from the facing page (offset), or they conformto the shape of the body of the printed text(shadows). They usually appear yellowish in visible

13. Foxing, page 14

light but differ under UV. Offsets fluoresce paleblue-white to white; shadows fluoresce yellow-orangearound the body of the text. The condensationexplanation for browning proposes the same causes forthese types of staining as for foxing i.e. moisture.

13.2.4 Methods of Identification

A. Visual Examination in Normal LightTo determine shape, color, and quantity. Todetermine if foxing spots correspond on recto andverso.

B. Transmitted LightTo identify if there is a core or bullseye. Todetermine location and quantity of other spots orcores not visible in normal light viewing.

C. Microscopic ExaminationWill help identify smaller iron cores and determinetheir exact location in the paper matrix.

D. Ultra-violet Fluorescence (360 nm)This is probably the most useful tool conservatorshave at hand to examine foxing as it will identifythe presence of metal (bullseye) when this is notnecessarily clear with microscopic or visible lightviewing. Fluorescence is an indicator of permanentchange in the cellulose and may indicate degradationby metal-induced oxidation or fungal digestion.1. Yellowish-White Fluorescence: Areas which

fluoresce under UV but show no visible lightfoxing may indicate the early stages of growth.Brown staining develops later, spreading out fromthe center. More advanced foxing was described byPress as brown spots visible by white lightsurrounded by a clear rim of florescence.Scanning with a spectrofluorometer showed "twoclear florescence peaks, on each side of the brownspot, which lay exactly in the valley betweenthem" [Press 1976, 28].Areas which fluoresce under UV but show novisible-light foxing also may indicate a previousbleaching treatment. "After any browning has beenremoved by oxidizing bleaches...the lesions stilldiffered in color under ultraviolet light andusually had fluorescent patches" [Meynell andNewsam 1978, 467].

2. No Florescence: Iron particles do not fluoreseceand appear much darker than surrounding areas.See PCC 6.4.4.

13. Foxing, page 15

E. Spot TestsStain tests used in research permanently staincellulose and are therefore inappropriate for worksof art or archival collection material on paper. Inspot testing, the paper conservator is limited tothose cases where a sample can be removed from thework without detracting from the appearance andintegrity of the work of art or artifact.1. Acidity

pH can be tested with indicator strips or, moreaccurately, with a pH meter.

2. Sampling of bullseye to perform iron test:a. Potassium Ferrocyanide

Saturate the sample with dilute hydrochloricacid and then add a 1% solution of the reagent.A blue color indicates the presence of iron.This test is extremely sensitive and one canget a positive result with Whatman filter paperor plain blotters [RD]. See PCC 10.4.12.A.2.b.

b. Potassium SulfocyanideThe sample is allowed to absorb dilutehydrochloric acid almost to saturation. Adilute aqueous solution of the reagent is thenapplied with a dropper. A pink color shows thepresence of iron. This test for ferric iron issensitive to 1 ppm [Iiams and Beckwith, 1935,412]. This test is more sensitive than SEM-EDS(-1000 ppm) or XRF (-100 ppm) [RK].

c. Potassium ThiocyanateThe sample is tested with a 5% solution of thereagent to which one drop of 10N hydrochloricacid was added. A brown color, easilydistinguishable from the color of foxing,indicates the presence of iron [Iiams andBeckwith 1935, 412]. This test is sensitive to1 ppm.

3. Fungal ActivityCotton blue-lactophenol can be used foridentification of living tissue in paper samplesusing transmitted light microscopy. It may bepurchased ready-made from biology suppliers [RK].

F. Analytic InstrumentationEmployed in research to identify fungal presenceand/or metal ions.

13. Foxing, page 16

1. Scanning Electron Microscopy (SEM)Used by Cain and Miller to examine iron and coppercores in sampled foxing. They found that whilesome particles had irregular, bumpy surfaces,others had "nearly smooth surfaces and theirshapes suggested chips or flakes, perhaps fromwear in the paper-making machinery" [Cain andMiller 1982, 56-7]. Arai employed this techniquefor identification of fungi; it readily showshyphae and conidia of fungi, if present [Arai etal. 1988, 11].

2. Energy Dispersive X-ray Spectral Analysis (EDX orEDS)Using this technique, Cain and Miller identifiedcores of either iron or copper in all sampledfoxing [Cain and Miller 1982, 56]. As thisinstrument is not sensitive to lighter elementslike hydrogen and oxygen, it would not indicate ifthe iron was present in the form of hydrated ironoxides/hydroxides. Daniels and Meeks used EDX toidentify copper and related elements in foxing[Daniels and Meeks nd.,7]. Also used by Arai[1987, 1165].

3. X-Ray Fluorescence (XRF)Press used XRF and concluded that there was nodifference in iron concentration between foxed andunfoxed areas [Press 1976, 29]. Cain and Millerused a more sophisticated XRF set-up to locate aniron core in a foxing sample where SEM and EDXwere unsuccessful [Cain and Miller 1982, 57].

4. Flameless Atomic Absorption SpectroscopyTang found this to be a "simple, rapid, andextremely sensitive method of determining metalsin cellulosic materials directly on the solidsample." Most areas of foxing analyzed showedhigher iron or copper concentration thansurrounding unfoxed areas [Tang 1978, 28, 31].

5. X-radiographyUsing x-rays generated at a 10kV potential, metalparticles were visible on a radiography plate aswhite specs on a black background. Daniels andMeeks found that all visible foxing they examined"displayed a nucleus of a small particle ofradiographically dense material, but there weremore particles detected on the x-ray plate thanwere visible in the paper" [Daniels and Meeks nd.,5].

13. Foxing, page 17

6. Thin Layer Chromatography (TLC)Employed to detect oligosaccharides and aminoacids from foxed spots [Arai et al. 1988, 12].Also used to show that foxing contained many ofthe same degradation products as found in normalcellulose [Cain 1983, 16]. A later study by Cainused TLC separation of extracts of foxed spots andof cultures of the isolated fungus, A. repens,showed certain unique bands in addition to thosecharacteristic bands produced by extracts ofunfoxed areas of the same paper [Cain, Stanley andRoberts 1987, 24].

7. IsotachophoresisUsed to analyze organic acids [Arai et al. 1988,1166].

8. High-powered Fluorescent Microscopy (160-1000x)Used to examine fungal mycelium and hyphae[Meynell and Newsam 1978, 467].

13.2.5 Effects of FoxingA. Support

1. SizingMeynell and Newsam observed that stained areas wetmore easily than the rest of the sheet indicatingfungi may grow, thrive on, and destroy the size[Meynell and Newsam 1978, 467]. It is possiblethat certain fungi prefer size to cellulose andwill only attack cellulose when the available sizeis exhausted. The resulting uneven sizing inpaper could cause uneven washing and bleachingduring conservation treatments.

2. CelluloseCain found that the degradation products of foxingspots are similar to the degradation products ofcellulose and suggested that some cellulolyticactivity is taking place, not simply a degradationof the size [Cain 1983, 15]. As by-products ofdigestion, fungi will excrete a variety of acidsincluding malic, fumalic, lactic, and acetic, withresultant damage to cellulose [Arai 1987, 1166].Iiams and Beckwith found the tensile strength ofthe foxed areas was considerably less than theunaffected parts of the same sheet, and that foxedareas were always more acidic than unfoxed areas[Iiams and Beckwith 1935, 409].Research shows that metal ions catalyze cellulosedegradation, producing oxidized cellulose andlarge amounts of acid. The higher concentration

13. Foxing, page 18

of acidity in foxed areas also contributes tocellulose deterioration.

B. Media

1. PastelGum binder is a good nutrient; some colors (darkrather than light) are preferred due to higherbinder content [BF].

2. Color PrintsLoss of color in patterns which approximate foxingareas have been observed in the blues. It may beacid attack of ultramarine by negative or reversefoxing [JCW].

3. Miniatures on Paper and IvoryGum binder in watercolor may be attacked, possiblydue to the high proportion of hygroscopicadditives (glycerin, honey) [BF].

13.3 Materials and Equipment

13.3.1 Examination Equipment

A. Near Ultra Violet ("Black Light")Radiant energy in the range of 320 to 400 nanometers.

B. Spot Test Reagents

C. CulturesReady made,sterile nutrient pad kits are availablefrom Sybron/Nalge. They have a limited shelf life.

D. Stereomicroscope

13.3.2 Treatment Equipment

A. Chemical Reagents

1. Alkaline water2. Bleaching reagents3. Reducing agents4. Chelating agents5. Acids

B. Tools

1. Scalpel/Sampling blade

13. Foxing, page 19

2. Microscope slides3. Small brushes for localized work4. Vacuum suction table for localized work

13.4 Treatment Variations13.4.1 Environment

A. HousingThere is some experimental evidence that foxing willworsen over time if kept in a poor environment. Asfor any other damaged, brittle or inherently fragilematerials, proper housing with non-acidic or bufferedmaterials, non-damaging RH and temperatures andlimited handling and exposure must be considered thefirst treatment step, which may mitigate furtherdamage [Beckwith and Iiams 1935, 415-16].

B. Relative Humidity (RH)Storing paper at a low RH is recommended as "the bestprecaution against foxing" [Daniels 1988, 93]. See13.2 . 1.A.3.a and 13.2 . 1.B.1.d.

C. TemperatureFungi generally prefer temperatures of 25C to 35C,dependent on species. There is no indication thatiron corrosion is temperature dependent. Oftentemperature in storage or exhibition spaces isdetermined by comfort zones for people, and it istherefore easier to regulate RH to avoid conditionswhich may further damage paper.

D. VentilationGood circulation is often mentioned as a deterrent tomold and air borne fungi especially in articlesrelating to libraries [Allsop 1985, 533].

13.4.2 No InterventionWalsh rightly notes that "foxing patterns do not alwaysconstitute a disfigurement to a work of art." Sheindicates a charcoal-and-ink wash drawing by Daumier inwhich the artist executed the work on a sheet of foxedpaper and incorporated the snowflake pattern into hisdesign [Walsh 1985, 8]. Some papermaking historians feelthat metallic inclusions are part of the paper's historyand should not be removed [RF].Harding found that works of art on prepared paper (e.g.silverpoint) were "in general not susceptible to foxing,although one or two fox marks may occasionally be

13. Foxing, page 20

evident. It would be very unwise to attempt removal ofthe spots, as this would normally involve the localisedapplication of aqueous solutions... inevitablyresult[ing] in an unsightly spot with a surroundinghalo, and disruption of the reflective properties of thesurface of the drawing" [Harding 1986, 26].

13.4.3 FungicidesMost conservators do not use fungicides. Though avariety of substances have been used as fungicidalfumigants (thymol, ortho-phenylphenol, etc.), none areactually fungicides as such (though one specialistmaintains that ortho-phenylphenol is fungicidal [RK]).Rather, they are fungistatic, inhibiting fungal growthonly while the toxin is present. Dissipation orevaporation of the substance will allow fungal growth tocontinue. A true fungicide would kill all fungi andspores on contact [Haines and Kohler 1986, 50].

13.4.4 Alkaline Washes

A. Neutralization of AcidityAs mentioned above, both fungal action and metal ioncatalysis will result in acidic degradation products.To prevent further acid-catalyzed hydrolysis of paperthese should be neutralized.Beckwith et al. state that the intensity of color isgreater in an acidic environment than when a sheet isalkaline. This indicates that imparting an alkalinereserve during sheet formation might reduce theseverity of future discoloration [Beckwith et al.1940, 311].

B. Reduction of StainingBeckwith et al. determined that the content ofmaterial soluble in a weak base (4% aqueous solutionof ammonium hydroxide) increased markedly in foxedareas as compared to unfoxed, and that this materialis hygroscopic [Beckwith et al. 1940, 322]. This mayin part explain why foxed areas in paper absorbmoisture first and most completely when a sheet ismoistened. Gallo and Hey found that washing withalkaline water can attenuate, if not completelyremove, many foxing marks. Deacidification, withhalf-saturated calcium hydroxide solution, was evenmore satisfactory [Gallo and Hey 1988, 102].However, many conservators would consider the pH ofthis solution too high [AK]. Percentages as high asthe four percent used by Beckwith et al. are notrecommended for treating works of art on paper,though less concentrated alkaline solutions have beenfound useful for reducing foxing.

13. Foxing, page 21

C. Treatment of Metal IonsIt is believed that both calcium and magnesiumcompounds form stable complexes with the transitionmetals (eg. iron and copper) and by so doing "retardcellulose depolymerization by deactivating thetransition metal catalysis", it should be noted that"magnesium [carbonate] is a more effective stabilizerthan calcium carbonate" [Williams et al. 1977, 49;57-58]. Hey believes this statement to be somewhatmisleading and discusses reasons why calciumcarbonate is the better choice for depositing andalkaline reserve [Hey 1979, 78-9]. It is believedthat imparting an alkaline reserve in the paper willprevent metallic ions from catalyzing cellulosedegradation, though this may not prevent degradationand discoloration of the metal itself [Tang andTroyer 1981,44]. When the reserve has been used upby the paper, metallic catalyzed deterioration ofcellulose may resume [JK]. Tang and Troyer foundboth calcium hydroxide and magnesium bicarbonateineffective in removing copper from handmade papers.Only ammonium carbonate significantly reduced thecopper content.

D. Pre-Treatment for BleachingConservators often find alkaline washing of foxing(either localized on the suction table or byimmersion) a good pre-treatment for bleaching. Notonly does this sufficiently diminish foxing stains inmany cases, it may help to inactivate metal ions ifpresent. Alkaline washes are also used to raisepaper pH so that the bleach can be used in the leastdamaging pH range. If iron is present and hydrogenperoxide is the bleach of choice, this would be aprerequisite. Conservators have employed thefollowing solutions for alkaline washing ordeacidification: calcium or magnesium carbonate andbicarbonate, calcium or magnesium hydroxide, andammonium hydroxide.

13.4.5 Metal Removal or InactivationMetal related foxing, particularly that with a visiblemetal core, cannot be successfully inactivated bybleaching alone, though the staining may be reduced oreliminated. Conservation literature indicates that inorder to prevent future metal-catalyzed oxidation of thecellulose, the metal should be mechanically and/orchemically removed, or rendered inactive. In practise,few conservators are currently using chelating agents todo this. Another suggested measure to render metalsinactive is controlled environment (especially humidity)and proper housing.

13. Foxing, page 22

A. MechanicalIron particles in the paper can be physically removedwith a scalpel under a microscope. It has beenrecommended that such treatment should be followed bychemical treatment to remove any residual ironparticles lodged in the fibers and to remove anystaining that might have surrounded the particle. Ifany iron/metal is left in the fibers, it willeventually re-oxidize and the stain will reappear[Burgess 1988, 24].

B. ChemicalThere have been several suggestions for decolorizing,complexing, or chelating metal impurities in paper.1. Sodium Dithionite (Hydrosulfite)

A 2-10% aqueous solution will convert theinsoluble colored ferric ion (Fe+++) to the moresoluble colorless ferrous ion(Fe++). Subsequentwashing with a Fe++ specific chelating agent willremove the ferrous ions from the paper and shouldprevent color reversion [Burgess 1988, 24, See PCC19.3.3.B.] Burgess also notes that the colorlessferrous iron can be removed by simple waterwashing, a chelating agent,however, would speedthe process. If colorless ferrous salts areallowed to remain in the paper, they will "in timebe oxidized by atmospheric oxygen back to thecolored ferric compounds" [Burgess 1991, 33]."Sodium hydrosulphite appears to be a relativelysafe agent in that it does not substantiallyyellow paper, weaken it, or alter it's pH"[Hawlye, Kawai and Sergeant 1981, 21].See also PCC 19.3.3.B.

2. Chelating Agents: These have been usedextensively in the paper industry and onlyrecently has some research been directed towardstheir use in paper conservation [Burgess 1991].a. EDTA and Related Compounds

Ferric or ferrous ion specific chelating agentshave been successfully used to sequester andremove iron from paper, though generally inconjunction with a reducing agent. "Chelatingreactions work fastest when both the chelatingagent and the metal to be chelated are presentin solution...An extremely good way to usechelating agents to remove iron is to couplethem with another process which involves thereduction of the insoluble ferric ion to themore water soluble ferrous form. Theconclusion is that chelation as a single

13. Foxing, page 23

process will not always be an ideal way toapproach removal of iron stains. Even inrelatively concentrated solution, [ chelatingagents] do not seem to work very well...[dueto] the low solubility of iron oxides andhydroxides" [Burgess 1991, 39]. Furtherresearch in the efficacy of ferric reagentssuch as Fe3 Specific (Dow Chemical Corp) wouldbe beneficial.

b. Oxalic AcidThis was often recommended in the literaturefor removing iron stains, but it is highlyacidic and its use is not advised. See Hawley,Kawai and Sergeant 1981.

3. Alkaline treatment of metal ions (See 13.4.3.C.).4. Acid Solubilization

"Examples of acids which can solubilize iron areoxalic [see 13.4.4.B.2.b] or acetic acid. Homeremedies such as vinegar or lemon juice have alsobeen used in the past. Providing the acid isstrong enough, iron removal will be relativelyquick and complete. However, subjecting paperartefacts to pH between 0 and 4 can, and usuallywill, do considerable damage...Therefore, the useof acids to remove iron stains is now discouraged"[Burgess 1991, 39].

13.4.5 BleachingBleaching would not remove metal impurities and mayresult in local metal-catalyzed degradation. This mayexplain the reappearance of metal-related foxing afterbleaching treatment. The recurrence of stains afterbleaching treatment may also be due to improper use ofbleaching solutions, inadequate pre-treatment, orinadequate rinsing. While visible discoloration maydisappear as a result of bleaching, florescence under UVwill continue [Meynell and Newsam 1978, 467].Special attention should be taken when bleaching foxedareas, as the cellulose in these areas is generally moredegraded than in surrounding areas and will be furtherweakened by oxidative bleaches.A. Chlorine Dioxide

Treatment can be either gaseous or aqueous. It hasbeen found that this bleach will "often decolorizesevere foxing without as much degradation or colorreversion as the hypochlorites" [PCC 19.3.1.D].Burgess emphasizes the importance of washing aftertreatment, whether gas phase or immersion bleachinghas been employed, to remove bleaching residues and

13. Foxing, page 24

solubilized discoloration products [Burgess 1988,22]. This bleach is the least affected by tracemetal elements in the paper.Meynell found that Cl02 was successful in reducingfoxing, though he was often left with a "lightbiscuit-coloured patch" rather than a totaldisappearance of the stain [Meynell 1979, 31].See also PCC 19.3.1.D.

B. Hydrogen PeroxideThe degradation of paper can be excessive ifunstabilized hydrogen peroxide is used, particularlywhen metal-induced foxing is the cause of thestaining being bleached. Stabilizers in the hydrogenperoxide will help prevent metal catalyzed breakdownof the bleach solution and therefore lessen cellulosedegradation during the procedure. Stabilizers suchas magnesium silicates and hydroxides used inhydrogen peroxide function similar to chelatingagents in that they act to complex metals and preventthem from participating in chemical reactions[Burgess 1991, 42]. Probably even stabilizedhydrogen peroxide should not be used on paper withconspicuous stains of heavy metals (e.g. iron,copper) or their salts or oxides unless complexing ofthese metals within the paper has also beenundertaken.This bleach is commonly used by conservators forfoxing stains, though most do warn against it if theybelieve iron is present. Presumably, this assessmentwould be based solely on examination by visible ortransmitted light, as most conservators reported thatthey rarely, if ever, used UV when examining foxing.The research has shown that all bullseye foxing iscentered on iron particles, and that snowflake foxinggenerally shows higher iron concentration thansurrounding areas [Cain and Miller 1982, 56; 60].This would contraindicate the use of hydrogenperoxide unless it could be shown that specificfoxing did not have a high iron concentration or ifiron ions could be inactivated or removed prior tobleaching.See also PCC 19.3.1.A.

C. Calcium HypochloriteGallo and Hey found that 5% solution was moreeffective in removing foxing stains than were 5%solutions of Chloramine-T and hydrogen peroxide(though these concentrations were not recommended for

13. Foxing, page 25

conservation use), and that reversion did not occurwith sheets rinsed in alkaline water [Gallo and Hey1988, 102]. Burgess suggests a much weaker bleachingsolution (0.1 to 0.5%) and recommends that calciumhypocholorite only be used on severely stainedartifacts in otherwise very good condition [Burgess1988, 22].Calcium hypochlorite is occasionally used byconservators for stubborn foxing stains and has beenfound to be especially effective. It has the addedadvantage of being a fungicide for some species offungi.See also PCC 19.3.1.B.1.

D. Chloramine-TA 2% solution was recommended [Coleman et al. 1969,197], though this bleach has generally fallen fromfavour due to the difficulty in removing residuesfrom the paper [Burgess 1988, 22].See also PCC 19.3.1.C.

E. Sodium BorohydrideThis moderate to weak reducing bleach generallyremoves mild stains but is only capable of partiallyeliminating severe staining. It is commonly used byconservators in concentrations of .1 to 2%, sometimesin combination with hydrogen peroxide treatment[Burgess 1988, 22].See also PCC 19.3.3.A.

F. Natural or Artificial LightUltraviolet radiation in the light source, combinedwith moisture in the paper, will produce peroxideswhich could then destroy mold (if present) and bleachdiscolorations [Hey 1983, 342]. This is best notused with papers with conspicuous stains of heavymetals (e.g. iron, copper) or their salts or oxides.This bleaching method is commonly used byconservators for foxing stains. Some work has beendone with localized light bleaching using a fiberoptic light source directed only at the foxing stainto be bleached.See also PCC 19.3.4.A and B.

13.4.7 Conservators' Observations of FoxingOn particular papers or artists' work that show anyunusual or consistent foxing patterns:

13. Foxing, page 26

A. Early Papers and Artists1. Many of Leonardo's and Durer's drawings and prints

are foxed. In a Durer print, The Triumph of Maximillian, executed on a composite support ofseveral pieces of paper joined together, foxingappears on only some of the papers. Foxing hasbeen observed on old Chinese papers and incunables[MH]

2. Have not seen much foxing on papers earlier than17th century, especially Italian [BF].

B. 17th and 18th Century Papers and Artists1. Tiepolo used what appears to be good quality laid

paper (watermarked with three crescents). Manyare heavily foxed overall [HO].

2. Have often found that with 17th/18th century ragpapers alkaline washing alone is sufficient toremove/reduce foxing so that bleaching is notnecessary. This is generally not observed to thebe case with 19th/20th century papers [RF].

3. Foxing of 18th and 19th century watercolors ongood quality papers often migrates from poorquality backing boards [JD].

C. 19th Century Papers and Artists1. Goya's complete set Disasters of War on antique

laid [BF].2. Mounted albumen prints foxed on photo but little

on mounts, or not in area of solid printed border[BF].

3. Matisse Jazz suite: small scattered bullseyefoxing [BF].

4. Arches wove [KGE].5. White spots on VGZ papers, some late 19th century

wove papers with lightly calendered surfaces [MS,see also 13.6 Special Considerations].

D. 20th century Papers and Artists1. Van Gelder Zonen - Picasso, Braque, Miro exhibit

snowflake and reverse foxing [BF; see SpecialConsiderations 13.6].

13. Foxing, page 27

2. Iron introduced into paper in solution: water usedin plein air watercolor painting. Dramaticoccurrence of foxing only on the part of the sheetthat was colored with wash. Artist unidentified,20th century American [NR]. It is not certainthat this is iron induced foxing or due to thepaper being damp.

3. Very rare to see foxing in gelatin silver photosof the 20th century - perhaps they are protectedby baryta layer or clay fillers in paper [BF].

E. Printing Papers

1. Printing papers which have been kept moist forlong periods of time before actual printing oftenshow foxing [PV].

2. Printmakers often observe fungal growth beforeprinting if they have not kept damp paperscarefully. The fungal activity shows up as spotswhich wet out differently from the rest of thesheet [SB].

3. Foxing seems more common in late 19th centuryAmerican printing papers rather than drawingpapers of the same time period. Snowflake foxingis often seen on unbound Audubon/Havell printswhich are Whatman and Whatman Turkey Mill papers[RF].

F. Chine Coll papers1. In many coll prints foxing is seen in area of

adhesive, as the paste is a good nutrient. Itwould be interesting to see if contemporary collprints using CMC will also become foxed whereadhesive is used [BF].

2. Redon chine con prints are often foxed [MHE].3. Heavy foxing noted on plate papers of many chine

coll prints where thin image bearing sheet has nostains [NH, SB].

4. The poor quality, hygroscopic plate paper oftenshows foxing [SB].

G. Japanese Papers

1. Have rarely seen foxing on Japanese prints.However, when observed the papers were oftenmounted to poor quality Western materials [JCW].

13. Foxing, page 28

2. Have often seen foxing on Japanese prints,especially those kept in Japan. There are certaintypes of traditional handmade papers such astengujo, which is force-dried on iron steamdriers. These show tiny scattered brown spotsbelieved (by Japanese) to be iron contamination[BF].

H. Effects of Light1. Have observed foxing in paper in areas in contact

with acidic poor quality window mats and none inexposed area. Uncertain whether this has to dowith light inhibiting the formation or whetherhigher humidity/acidity levels in covered areasare favoring formation [KGE].

2. Exhibition under fluorescent light seems to have"bleached foxing in 2-3 known cases" [KDL].

3. Have observed snowflake foxing which was"negative" (or lighter than the paper) in the areabeneath an acidic matboard window and "positive"(brown or darker) in the area exposed to light.Was its growth under the mat discouraged by alumrosin sizing in the mat? [JCW] Same phenomenonalso noted in reverse foxing on Van Gelder Zonenpapers where portion exposed to light by windowmount is dark brown [DDM]. See also 13.6 SpecialConsiderations.

13.4.8 Observations on Current Conservation PracticesThe following summarizes responses to the foxingquestionnaire as well as statements made bycontributors.A. Examination

1. Respondents are divided as to the cause of foxing,but many believe a combination of factors areresponsible.

2. Respondents easily differentiated fuzzy moldgrowth from foxing stains.

3. Respondents usually did not describe foxing byshape, fluorescence (or lack thereof), color orsize, although some did use the terms "bullseye"and "snowflake".

4. Respondents usually do not use ultra-violetradiation in the examination of foxing spots.

5. Respondents usually do not employ spot tests toconfirm the presence of iron/metal.

13. Foxing, page 29

B. Treatment1. Respondents do not often mechanically or

chemically attempt to remove metal from paper,although several respondents mentionedoccasionally trying acids and chelating agents.The following are examples:a. On iron spots have excavated centers. Have

also tried (without success) oxalic acid,hydrofluoric acid and titanium trichloride[BF].

b. Very dilute oxalic acid used to remove pin-point foxing on the suction table. Also havereduced dark bullseye foxing stains with dilutecitric acid [JCW].

c. Have used oxalic acid and EDTA on ironstains/specks found in foxing. The resultswere questionable and the aesthetic improvementwas minimal. Admittedly I have not triedthese chemicals in some time and it would seemtheir application has merit [NH].

d. If iron is present I'd perhaps try Versene FE3Specific,

13. Foxing, page 30

c. Wash art in alkaline water and bicarbonatesolutions. Have used calcium hydroxide,magnesium carbonate, magnesium, bicarbonate,and ammonium hydroxide solutions. If furtherremoval of foxing stains is needed, lightbleaching in an alkaline bath is preferredmethod [NH].

d. Have reduce bullseye foxing spots on the sky ofa watercolor on Whatman paper by locallywashing with dilute ammonium hydroxide on thesuction table [JCW].

e. It should be noted that significant improvementcan be brought about with ammonium hydroxideand this pre-treatment makes bleaching withhydrogen peroxide and sodium borohydride moreeffective [MHE].

f. I have had some success in removing/reducingfoxing with local application of ammonia usingsmall cotton poultices soaked in ammonia andplaced in the foxed areas. With many 17th and18th century rag papers I have found alkalinewashing is often sufficient to remove/reducefoxing so that bleaching in not necessary.Ammonia (pH 8.5) is also used as a pre-treatment when bleaching is necessary. Theammonia is applied locally with a brush andarea partially dried before immersing in waterfor sun bleaching [RF].

3. In bleaching foxing, conservators will almostalways apply bleach locally to spots first,followed either by an overall bleaching or byrinsing alone. The following are examples:a. For local applications use sodium borohydride

.5-2.0% with a couple drops calcium hydroxideadded to maintain pH, pre-wet areas withammonium hydroxide (pH 9) and local rinse withsame. Does not cause halos. Do not usehydrogen peroxide if possibility of iron(bullseye) is apparent. Also, it tends tocause halos [BF].

b. Sodium borohydride .5% and .1% solutions wereapplied locally to foxed areas of a pre-washedpencil drawing on paper. As the foxed spotswere already quite damaged a reducing bleachrather than an oxidizing bleach was preferred.Application of the the borohydride wasrestrained as it was anticipated the elevatedpH of the borohydride solution would also swell

13. Foxing, page 31

paper fibers locally and result in furtherrelease of color bodies in the rinse bath.Water rinsing resulted in further diminishingof foxed areas to a pale tan color which nolonger competed with the visual impact of thedrawing. Overall treatment of drawing in bleachwas not considered appropriate, both because ofthe risk of paper delaminating in smallblisters and because the unfoxed portions ofthe paper showed a tendency to lightenconsiderably when spot tested [KN].

c. Have seen red-brown foxing reduced with sodiumborohydride go light gray-looking dirty. Thiscould not be oxidized back to white withperoxide [RF].

d. If a hard sized paper is foxed, I willsometimes float the sheet on a bath ofperoxide. The bleach comes up through thespots but not through the rest of the sheet[JCW].

e. Follow alkaline washing and drying of the sheetwith local application to spots of eitherhydrogen peroxide (3%, pH raised to 9.0-9.5with calcium hydroxide) or sodium borohydride(.5%). Allow to dry. Repeat as necessary.Minor rings which may form and remaining foxingmarks tend to bleach out when followed byoverall light bleaching. Rinse thoroughly inwater baths [NH].

4. Respondents most often mentioned using hydrogenperoxide as bleach of choice on foxing. Sunbleaching in water with calcium hydroxide addedwas also frequently cited. Reducing the foxingwith sodium borohydride was the third mostmentioned bleach. One respondent mentioned usingchlorine bleaches when hydrogen peroxide could notbe used due to presence of metal in paper. Thefollowing are examples:a. Use 2% hydrogen peroxide (pH 8.5 with ammonium

hydroxide and calcium hydroxide added). Use 3-4 applications drying between each application.Use hydrogen peroxide only until slightbubbling begins, then discard and mix newsolution if necessary [SB].

b. Sunlight or sunlamps are used rather than afluorescent light bank. Adding a few drops ofhydrogen peroxide to the tray seems to catalyzethe effect of light [NH].

13. Foxing, page 32

c. Have used hydrogen peroxide in water-ethanolmixes on the suction table for charcoal orwhite chalk drawings which were badly foxed.Have sun bleached a foxed intaglio image byAlbers in 75% ethanol, then washed in the samemix to great effect [JCW].

d. For artwork which cannot be immersed, wash artas possible (suction table, disk, or byfloating) and light bleach by local wetting offoxed areas and exposing to light source. Masksensitive areas. Rinse thoroughly as above[NH]. One technique for shorter light.bleaching time is to locally bleach foxed spotswith hydrogen peroxide, rinse, then place insunlight bath. Conservators sometimes add afew drops of hydrogen peroxide to the bath usedin light bleaching.

5. Some respondents noted that in some cases foxingcould not be completely diminished and that theyaccepted a pale ghost-image of the stain ratherthan continue bleaching.

6. Respondents generally have noted few incidences ofcolor reversion. Some noted reversion withhydrogen peroxide, but often attributed this topoor rinsing.

7. Some respondents stated they believe that chlorinebleaches to be useful on fungal foxing.

8. Respondents generally do not use a fungicide forfoxing.

13. Foxing, page 33

13 . 5 Bibliography

AIC-BPG. Paper Conservation Catalogue - 10. Spot Tests. Washington,D.C.: AIC-BPG, 1990.AIC-BPG. Paper Conservation Catalogue - 19. Bleaching. Washington,D.C.: AIC-BPG, 1989.Allsop, Dennis. "Biology and Growth Requirements of Moulds and OtherDeteriogenic Fungi." Journal of the Society of Archivists 7:8, 1985, pp.530-533.Ambler, H. R. and C. F. Finney. "Brown Stains Formed on Wet Cellulose."Nature 179, 1957, pp. 1141.Arai, Hideo. "On the Foxing-Causing Fungi." ICOM Committee forConservation: 8th Triennial Meeting, Sydney, Australia, 6-11 September, 1987, Preprints Marina del Rey: The Getty Conservation Institute, 1987,pp. 1165-1167.Arai, Hideo, Noritaka Matsumura and Hiroyuki Marakita. "Induced Foxingby Components Found in Foxed Areas." Preprints from the Ninth TriennialICON Meeting Dresden, 1990, pp. 801-803.Arai, H., N. Matsui, N. Matsumura and H. Murakita. "BiochemicalInvestigations on the Formation Mechanisms of Foxing." The Conservationof Far Eastern Art: Preprints of the Contributions to the KyotoCongress, 19-23 September, 1988 London: IIC, 1988, pp. 11-12.Baynes-Cope, D. "Some Observations of Foxing at the British MuseumResearch Laboratory." Ind. Biodeterioration Bulletin 12:1, 1976, pp. 31-33.

Beckwith, T. D., W. H. Swanson and T. M. Iiams. "Deterioration ofPaper: the Cause and Effect of Foxing." Publications of the Universityof California at Los Angeles in Biological Sciences 1:13, 1940, pp. 299-356.Burgess, Helen D. "Practical Considerations for ConservationBleaching." Journal of the International Institute for Conservation -Canadian Group 13, 1988, pp. 11-26.Burgess, Helen. "The Use of Chelating Agents in ConservationTreatments." The Paper Conservator 15, 1991, 36-44.Cain, C. Eugene. "Evidence for Multiple Causes of Foxing in Paper." AICPreprints of Papers Presented at the 16th Annual Meeting, New Orleans: June 1-5 1988, Book and Paper Group Abstracts. pp. 275-76.Cain, C. Eugene and Victor F. Kalasinski. "Characterization ofEighteenth- and Neneteenth-Century Papers Using Fourier TransformInfrared Spectroscopy and Thin-Layer Chromatography", Application of

13. Foxing, page 34

Science in Examination of Works of Art: Proceedings of the SeminarSeptember 7-9, 1983, Museum of Fine Arts, Boston (1987), pp.55-58.Cain, C. Eugene and Barbara A. Miller. "Photographic, Spectral andChromatographic Searches into the Nature of Foxing." AIC Preprints of Papers Presented at the 10th Annual Meeting, Milwaukee, Wisconsin, May26-30, 1982 Washington: AIC, 1982, pp. 54-62.Cain, C. Eugene and B. A. Miller. "Some Examples of Inorganic Foxing",IIC 9th International Congress, Abstracts of Poster Sessions,Washington, D.C., Septermber 3-9, 1982, p.7Cain, C. Eugene, M. B. Stanley, and W. H. Roberts. "Paper Foxing:Biochemical Effects of Fungal Infections of Paper", Journal of theMississippi Academy of Science, Vol 32, 1987, p.24.Cain, Eugene. "The Analysis of Degradation Products Extracted fromSelected 19th Century Papers." The Book and Paper Group Annual 2, 1983,pp. 15-18.Cain, Eugene and Barbara A. Miller. "Proposed Classification of Foxing."Postprints from the AIC 10th Annual Meeting, Milwaukee, Wisconsin, May26-30, 1982 Washington: AIC, 1984, pp. 29-30.Carter, John. ABC for Book Collectors. New York: Granada, 1980.Coleman, M. H., A. D. Baynes-Cope and R. C. Agabeg. "Foxing." Chemistryand Industry 7, 1969, p. 197.Daniels, V. and N. D. Meeks. Foxing Caused by Copper Alloy Inclusions in Paper. Department of Conservation, The British Museum, n d. [Also tobe published in Symposium 88: Conservation of Historic and ArtisticWorks of Art Canadian Conservation Institute, Ottawa.]Daniels, Vincent. "The Discoloration of Paper on Ageing." The PaperConservator 12, 1988, pp. 93-100.Danzig, Rachel. Past and Current Research into the Causes of Foxing onPaper. Unpublished paper submitted to Robert Koestler at theConservation Center, Institute of Fine Arts, New York University,December 14, 1991.Engelbrecht, Gerda. The conservation of a watercolour by P. Wilson Steer entitled "Under the Cliffs, Dover", with art historical research, media analysis and an investigation into the penomenon known as foxing.A special project presented at Newcastle Polytechnic for the Degree ofMaster of Arts in Conservation of Fine Art, April 1991 (unpublished).Feniak, Christine J. "Treatment of a Parasol Using a Reductive Bleach."Journal of the International Institute for Conservation, Canadian Group6, 1981, 31-34.

13. Foxing, page 35

Gallo, Fausta. "Biological Agents which Damage Paper Materials inLibraries and Archives." Recent Advances in Conservation: Contributions to the IIC Rome Conference, 1961. London: Butterworths,1963, pp. 55-61.Gallo, Fausta and Margaret Hey. "Foxing - A New Approach." The PaperConservator 12, 1988, pp 101-102.Haines, John H. and Stuart A. Kohler. "An Evaluation of Ortho-phenylphenol as a Fungicidal Fumigant for Archives and Libraries."Journal of the American Institute for Conservation 25, 1986, pp. 49-55.Harding, Eric. "Restoration of Drawings Executed on Prepared Paper."The Conservator 7, 1983, 24-28.Hawley, Janet K, Elizabeth A. Kawai and Christopher Sergeant. "TheRemoval of Rust Stains from Artic Tin Can Labels Using SodiumHydrosulphite." Journal of the International Institute for Conservation, Canadian Group 6, 1981, 17-24.Hey, Margaret. "Foxing: Some Unanswered Questions." The Antiquarian BookMonthly Review 10:9, 1983, pp. 340-343.Hey, Margaret. "The Washing and Aqueous Deacidification of Paper." ThePaper Conservator 4, 1979, pp. 66-80.Hunter, Dard. Papermaking: The History and Technique of an AncientCraft. New York: Dover Publications, Inc., 1943.Hutchins, J. K. "Water Stained Cellulosics: A Literature Review."Journal of the American Institute for Conservation 22:2, 1983, pp. 57-60.Iiams, Thomas M. and T. D. Beckwith. "Notes on the Causes and Preventionof Foxing in Books." The Library Quarterly 5:1, 1935, pp. 407-418.Ligterink, Frank J., Henk J. Porek and Wim J. Th. Smit. "Foxing Stainsand Discoloration of Leaf Margins and Paper Surrounding Printing Ink:Elements of a Complex Phenomenon in Books." The Paper Conservator 15,1991, 45-52.Meynell, G. G. and R. J. Newsam. "Foxing, a fungal infection of paper."Nature 274:5670, 1978, pp.466-468.Meynell, Guy. "Notes on Foxing, Chlorine Dioxide Bleaching andPigments." The Paper Conservator 4, 1979, pp. 30-32.Meynell, Guy and Ray Newsam. "Foxed Paper and Its Problems." NewScientist 1979, p. 567.

13. Foxing, page 36

Nol, L., Y. Henis and R. G. Kenneth. "Biological Factors of Foxing inPostage Stamp Paper." International Biodeterioration Bulletin 19:1,1983, pp.19-25.Press, R. E. "Observations on the Foxing of Paper." InternationalBiodeterioration Bulletin 12:1, 1976, pp. 27-30.Se, P. Les Maladies du Papier Piqu. Paris: 0. Dain et Fils, 1919.Sharpely, J. M. and M. E. King, "Laboratory Analysis of Problems inPapermakers Felts" Biodeterioiration of Materials, vol 2, A. HarryWalters and E.H. Hueck-van Der Plas, eds. New York: John Wiley & Sons,1971, 1161-1167.Tang, Lucia. "Determination of Iron and Copper Content in 18th and 19thCentury Books by Flameless Atomic Absorption Spectroscopy." Journal of American Institute for Conservation 17:2, 1978, pp. 19-32.Tang, Lucia and Margaret Ann Troyer. "Flameless Atomic AbsorptionSpectroscopy: a Useful Tool for Direct Determination of Elements inLibrary/Archival Materials." Technology and Conservation 6:2, 1981, pp.40-45.Walsh, Judith. "Selected Conservation Problems." Drawing 7:l, 1985, pp.5-8.Williams, J. C., C. S. Fowler, M. S. Lyon and T. L. Merrill. "MetallicCatalysts in the Oxidative Degradation of Paper." Preservation of Paperand Textiles of Historic and Artistic Value, John C. Williams, editor(Advances in Chemistry Series No. 164). Washington, D. C.: AmericanChemical Society, 1977, 37-61.

13. Foxing, page 37

13.6 Special Considerations: Reverse Foxing

13.6.1 DefinitionThis is an informal term coined by conservators whonoticed spots in paper which had a much lighter tonethan the surrounding sheet. This is sometimes referredto as negative foxing. The phenomenon may not actuallybe foxing, i.e. it may not be related to metallic orbiological causes. These spots do, however, have someof the characteristics described in Cain'sclassification of foxing and, for lack of a better term,at present these stains or spots will be referred to asreverse foxing. There has been little or no research onreverse foxing. Observations have been made of numerousreverse foxing spots on Van Gelder Zonen (VGZ) papers,in particular those used by Vollard to print Picasso'sSaltimbanque series. These are on wove paper with VANGLEDER ZONEN watermark. Reverse foxing is not isolatedto VGZ papers but has been noted by conservators onArches and Ingres papers as well.There are several Van Gelder Zonen papers with differentwatermarks [DDM]. Those observed are:A. Fleur-de-lys with VGZ/ZB. VAN GELDER ZONEN/HOLLANDC. VAN GELDER ZONENWorks by Picasso, Braque, and Miro exhibit snowflake andreverse foxing [BF].Works by Pennell, Zorn, and Shaw also exhibit snowflakeand reverse foxing [DDM].

13.6.2 DescriptionThere were three types of spots observed on the VGZpapers examined. Whether these spots are caused byuneven sizing or fillers in the sheet, or perhaps by apH sensitive dyes is not known. Whether the three spotsnoted are different stages of the same thing is notknown. For conservators VGZ papers, and perhaps otherswith reverse foxing, can often be problematic (see13.6.3.) [TO].A. Visible white splotches similar to "snowflake" which

are lighter in tone than the sheet. They are ofvarying size but generally quite large, round, andhave uneven diffuse edges. In some cases the whitesplotches have tiny whiter cores. They are locatedalmost invariably within the plate mark area on theverso of the paper and are rarely seen on the recto,except sometimes as small pinpoint white spots which

13. Foxing, page 38

correspond to centers of the splotches verso. Thesespots generally fluoresce a brighter white.

B. Visible white circular spots lighter in tone than thesheet. These are generally quite small, have a hardsharp edge and are distinctly round. These spots arenot restricted in location and were found on therecto, verso, plate and in the margins. Often thesespots seem to migrate through the paper. Thesefluoresce white.

C. Not visible, or only faintly visible, in normallight, these small greyish/brown spots or inclusionare noted in transmitted light and UV. They do notfluoresce, appearing blue/black under UV. Theyappear to have a core.

13.6.3 Treatment WarningsCaution should be taken when attempting to bleach VanGelder Zonen papers which have reverse foxing.Numerous conservators have found areas of the paper notpreviously white will turn stark white during bleaching.It is possible the areas which turn stark white duringbleaching are the invisible brown inclusions found underUV examination. Some conservators have stated they gotstark white spots with water washing alone! Bleachesreported to cause this to occur are: hydrogen peroxide,sun bleaching, sodium borohydride, Chloramine-T andcalcium hypochlorite. One bleach which has beenreported to have some success is chlorine dioxide, usedat approximately .2% for only several minutes. Thisbleach can also cause white spots but the short bleachtime seems to allow conservators to reduce stains asnecessary and begin rinsing before the stark white spotscan begin to appear [TO].

13. Foxing, page 39

Participants (listed alphabetically)Compilers: Jonathan Derow and Antoinette OwenContributors: Sarah Bertalan, C. Eugene Cain, Rachel Danzing, AnneDriesse, Katherine Eirk, Margaret Holben Ellis, Reba Fishman, BettyFiske, Patricia Dacus Hamm, Nancy Heugh, Margaret Hey, Antoinette King,Robert J. Koestler, John Krill, Kendra D. Lovette, Deborah D. Mayer,Elizabeth Morse, Kitty Nicholson, Nina Rayer, Sarah C. Riley, PaulaVolent, Judy WalshQuestionnaire Respondents: Katherine Eirk, Margaret Holben Ellis, BettyFiske, Nancy Heugh, Margaret Lawson, Elizabeth Morse, Helen Otis,Marjorie Shelley and Paula Volent.Editorial Board Liaison: Sarah BertalanEditorial Board: Sylvia Rodgers Albro, Sarah Bertalan, Antoinette Dwan,Catherine I. Maynor, Catherine Nicholson, Kimberly Schenck, Ann Seibert,Dianne van der Reyden, Terry Boone Wallis

2/13/92

Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9Page 10Page 11Page 12Page 13Page 14Page 15Page 16Page 17Page 18Page 19Page 20Page 21Page 22Page 23Page 24Page 25Page 26Page 27Page 28Page 29Page 30Page 31Page 32Page 33Page 34Page 35Page 36Page 37Page 38Page 39Page 40Page 41Page 42