http://kulturarvsdata.se/raa/fornvannen/html/2016_017 Fornvännen 2016(111):1 s. 17-26
Ingår i samla.raa.se
Pigment traces on medieval stonework in Gotland’s churches – examination of seven 12th century baptismal fonts and a limestone pew
By Anders G. Nord, Kate Tronner, Kjell Billström and Marianne Gustafsson Belzacq
Nord, A.G., Tronner, K., Billström, K. & Gustafsson Belzacq, M., 2016. Pigment traces on medieval stonework in Gotland's churches – examination of seven 12th century baptismal fonts and a limestone pew. Fornvännen 111. Stockholm.
The authors have analysed pigments on eight medieval stone objects in Gotland’s churches. They are seven 12th century baptismal fonts made of sandstone and a ce
remonial limestone pew in Burs church from about 1350. Fragments of medieval paint were sampled and analysed by SEM/EDX, FTIR, GCMS and MCICPMS.
57 samples were collected. Infinitesimal traces of cinnabar, ultramarine and gold foil were sparsely found. Iron oxide (Fe2O3) and lead pigments occur commonly, while green and blue pigments seem less frequent. Later “improvements” of green copper arsenites were found on the limestone pew in Burs church. Lime has prima
rily been used as a binding medium, possibly in combination with now complete
ly degraded organic substances. Isotope analyses indicate that the lead pigments most likely have a provenance in the Harz and Erzgebirge mountains in Germany, possibly also in Eisleben. Other pigments have more remote origins, which reflects Visby’s significance as a trade center.
Anders G. Nord, Kevingeringen 10, SE–182 50 Danderyd andersgn@tele2.se
Kate Tronner, Flädergränd 2, SE–187 73 Täby katetronner@gmail.com
Kjell Billström, Department of Geosciences, Swedish Museum of Natural History, Box 50007, SE–104 05 Stockholm
kjell.billstrom@nrm.se
Marianne Gustafsson Belzacq, Gurpe 117 Kräklingbo, SE–623 70 Katthammarsvik marianne.g.belzacq@gmail.com
Hundreds of medieval churches in Sweden have more or less wellpreserved murals painted al secco on lime ground. We have earlier documented and analysed many of their pigments (Nord et al. 1996;
2000; 2011; 2015), and have so far identified 23 pigments (disregarding modern ones used for later
“improvements”). Recently, we have extended our studies to include medieval stonework such as outdoor portals (Nord & Tronner 2014). For the
present study, we analysed original paint traces
from seven 12th century baptismal fonts made
from sandstone and a ceremonial limestone pew
from around 1350, all in Gotland’s churches, by
means of SEM/EDX and other techniques. The
aim of the study has been to identify pigments of
various shades and search for binding media. While
the stone itself is obviously of local origin, we have
tried to identify the origin of the pigments.
Gotland has 92 wellpreserved medieval stone churches, in addition to ruined ones. Most of these churches have a medieval baptismal font, made from sandstone or limestone from local quarries.
Most fonts have been painted, and original paint can still be observed as infinitesimal traces. How
ever, many fonts were redecorated 300–400 years ago, which strongly reduced the number of usable objects for the present study.
Nearly a century ago, art historian Johnny Roosval (1918; 1925) identified a number of sculp
tors or workshops, and named them Byzantios, Majestatis, Hegvald, Calcarius, Sigraf, Egypticus etc. Many of these artisans seem to have worked on Lund Cathedral before moving to the pros
perous island of Gotland to work there. In his 2012 survey of Gotland fonts, Svenrobert Lund quist attributes each of 45 baptismal fonts to a specific group based on type, motif, decorations, reliefs, chisels used, etc. These fonts include the ones selected for the present study, and we use the attributions suggested by Roosval and Lund
quist. It should also be added that the influential Byzantine art of the same era in Sweden has been discussed by for instance Andreas Lindblom &
Gunnar Svahnström (1959), Erland Lagerlöf (1999) and Svetlana Vasilyeva (2009).
Among the sculptors, the Byzantios group
seems to have been the most productive. Roos
val’s invented name Byzantios suggests a Russian
Byzantine origin, but this has so far not been generally accepted by art historians. Presumably the name was suggested (Roosval 1918) by a cer
tain stylistic elegance influenced by Byzantine murals. Sixteen fonts on Gotland, some of them in poor condition, have been attributed to Byzan
tios. For the present study, the Hejde and Hogrän fonts were selected, rather from the presence of usable original paint than as typical representa
tives of the Byzantios group. Another sculptor named Hegvald or Hegwaldr was in all likeli
hood a local man from Gotland. His workshop has a more popular and provincial style, and three fonts were selected for this study, namely those of Endre, Etelhem and Stånga. The Etel
hem font is actually signed “Hegwaldr” in runes.
The Majestatis group, famous for their extreme skill in reliefs, is represented here by the font of Gerum. Roosval assigned this name because of the group's habit of representing Christ as sover
eign ruler of the world (Majestatis domini). Final
ly, the Master of Barlingbo is only known to have made the font in Barlingbo church (fig. 1). This beautiful and imposing font is the largest one on the island.
All seven selected fonts have beautiful sculp
tural decoration. The local sandstone they are made of consists mainly of quartz grains cemented to
gether by calcium carbonate. The ceremonial pew in Burs church however is made of local fine
grained limestone. It is probably a product of the Egypticus workshop. All paint traces for the study were carefully sampled and documented. Many pigments could not be observed with the naked eye, but demanded a magnifying glass. When
ever possible, we took somewhat larger samples with the aim to search for binding media. In total we collected 57 samples.
18 Anders G. Nord et al.
Fig. 1. The sandstone font in Barlingbo church, Got
land, with a sculpture of the archangel Gabriel. Pho
to: Bengt A. Lundberg, National Heritage Board of Sweden.
Pigment traces on medieval stonework in Gotland’s churches 19 SEM/EDX analyses of the pigments
Work began some years earlier with Kate Tron
ner identifying the pigments on the Barlingbo font by means of wet chemistry and Xray pow
der diffraction (XRD). All other pigment sam
ples were analysed by SEM/EDX, i.e. with a scan
ning electron microscope (Hitachi S4300) equip
ped with a LINK/Oxford unit for Xray micro
analysis, at the Swedish Museum of Natural His
tory.
Tab. 1 lists the results of the pigment analy
ses. The pigments identified accord well with those found on medieval stone portals (Nord &
Tronner 2014), although gold and ultramarine were never found there. Tab. 2 offers further che
mical and mineralogical data. The pigment PbO is denoted as massicot, whether from the natural mineral or a synthetic product with that compo
sition. Furthermore, plattnerite (βPbO2) is not an original pigment but a product of the oxidation of a lead pigment. Chlorine was detected in many samples, probably representing remains of mod
ern cleaning agents. On the Burs pew, “improve
ments” of copper arsenites were found. These are likely to represent Schweinfurter green and Scheele’s green, as judged from the Cu:As atomic ratios.
Tab. 3 lists the pigments we found on the fonts and the pew. Note that pigments that are now missing from an individual object may orig
inally have been there. Iron oxide was found on most of the fonts, and so was minium and lead white. The coeval workshops Byzantios, Hegvald and Majestatis, to which the examined fonts have been attributed, seem to have used similar pa
lettes. Fragments of gold, cinnabar and ultrama
rine on some objects is consistent with the fact that Gotland was a prosperous island from the Iron Age through the Middle Ages. Note that generally green and blue pigments occur sparsely here compared to results obtained for medieval ecclesiastical stone objects on the Continent (e.g.
RossiManaresi 1981; 1984; Hauff 1988). On these stone objects, malachite, azurite, ultrama
rine and cobalt glass (smalt) are common.
The plattnerite problem
Let us look closer at plattnerite. Lead pigments painted on a ground of plaster (lime) often dark
en over time. Accordingly, lead white, massicot
and minium transforms to (usually) black platt
nerite (e.g. Giovannoni et al. 1990; Nord et al.
1996). This transformation may be partial or complete. In the former case the colour may become dark brown or a mixture of dark shades.
This process seems to be slow as any rapid change to black candle flames and haloes on the murals would have evoked a congregation’s righteous anger. The transformation process requires O2(g) from the air as oxidizing agent. Coauthor M.G.B., with a lifelong experience of medieval murals, has noted that minium can retain its red colour in cavities covered and protected by plaster.
We have also found that sometimes the oxi
dation does not take place in paintings on other substrates. A striking example is found in the church of Härkeberga in Uppland. In the narrow staircase leading to the organ loft, there is an image from around 1490 of a man, originally paint
ed with minium to wear orange clothes. Part of the painting is on lime ground where plattnerite has formed, another part on oak timber where the minium is preserved (Nord & Tronner 2000).
Moreover, lead pigments on wood from the (longsince demolished) medieval stave church of Dalhem on Gotland still preserve their original colour (Nord & Tronner 2011).
Our study of the baptismal fonts yielded a similar result. Plattnerite has almost always form
ed when lime is present (tab. 1). Thus it seems that lime, or some substance(s) therein, acts as a kind of catalyst for the oxidation process. The “platt
nerite problem” has been discussed by other authors. Petushkova & Lyalikova (1986) empha
sized the influence of microorganisms for the oxi
dation of lead pigments. Giovannoni et al. (1990) studied the oxidation process of lead white. They concluded that an alkaline environment, mois
ture and microorganisms all seem to accelerate the oxidation process. Aze et al. (2007; 2008) in dicate light, humidity, microorganisms and sul
phurous air pollutants as important factors for the formation of plattnerite. Kotunalová et al.
(2009) stress inorganic salt as one of the most active agents. To sum up, lead pigments on a lime ground exposed to air, in combination with oth
er damaging factors, seem to be susceptible to
plattnerite formation.
20 Anders G. Nord et al.
Church Workshop Sample Colour Pigments identified
Hejde
Hogrän
Endre
Etelhem
Stånga
Gerum
Barlingbo B
B
H
H
H
M
MB
10A 10B 11 12 13*
14*
5 6*
7*
8 9
1A 1B 2 3 4A 4B
15 16*
17 18 19 20
35*
36 37 38
28 29 30 31 32 33 34
MB1 MB3 MB4 MB6 MB8 MB9 MB10 MB11 MB12
Red Dark grey Reddish brown Orange Black Black
Yellow Dark grey Orange Violet Yellow
Black Red Yellow Dark grey Red Black
Reddish brown Red
Beige Yellow Blue Pink
Red Golden Bluegrey Reddish brown
Golden Pink Green Red Grey Grey to black Reddish brown
Red Blue Blue Red White Red Blue Red Pale green
Iron oxide, lead white Plattnerite, lime
Iron oxide, minium, plattnerite Iron oxide, massicot, lead white Plattnerite, lime
Plattnerite, lime
Massicot, ochre, minium, lead white Plattnerite, lime
Minium Caput mortuum Massicot
Plattnerite, lime Iron oxide, lime Orpigment Plattnerite Cinnabar Plattnerite, lime
Iron oxide Minium
Orpigment, lead white Orpigment
Azurite, lead white Minium with some lime
Iron oxide, minium Gold
Ultramarine, lead white, lime Iron oxide
Gold
Minium, lead white Malachite, lead white Cinnabar, minium Minium, plattnerite, lime Plattnerite, lime Iron oxide
Iron oxide, minium, gold Azurite
Azurite, lead white, carbon black Minium, lead white
Lime
Minium, lime, gold Azurite, lead white Minium, lead white Green earth
Pigment traces on medieval stonework in Gotland’s churches 21
Church Workshop Sample Colour Pigments identified
Burs (pew) E 21
22 23 24 25*
26 27
Bluegreen
Black
Brownish black Greyblack Black Bluegreen Green
Atacamite, malachite, azurite, copper arsenite (cf. tab. 2) Plattnerite, lime
Plattnerite, minium, lime Plattnerite, lime Plattnerite, lime Malachite, azurite Atacamite
Table 1. Pigments (including oxidized plattnerite, βPbO2) identified on seven baptismal fonts and a limestone pew. Abbreviations for the sculptors and the workshops (Workshop): B=Byzantios, H=Hegvald, M=Majestatis, MB=The Master of Barlingbo, E=Egypticus. The seven samples used for lead isotope analysis are marked with an asterisk (*).
Colour Name Chemical composition; mineralogical information
White
Black
Yellow
Red
Violet Green
Blue
Lime Lead white
Carbon black Plattnerite
Ochre Massicot Orpigment Gold Iron oxide Minium Cinnabar Caput mortuum Malachite Atacamite
Green earth Scheele’s green Schweinfurter green (many synonyms exist) Azurite
Ultramarine
Calcium carbonate, CaCO3
Synthetic product 2PbCO3.Pb(OH)2. Also exists as the mineral hydrocerussite
C
Lead dioxide, βPbO2 (oxidized form of a lead pigment, i.e. not a natural pigment)
Earth colour containing iron oxide(s)
The mineral PbO (also exists as a synthetic product) Mineral As2S3
Au (natural product)
Fe2O3 (natural or synthesized) Synthetic product Pb3O4 Mineral HgS
Violet modification of Fe2O3 Mineral CuCO3.Cu(OH)2
Very rare mineral. Usually a synthetic form of Cu2Cl(OH)3 has been used
Ironrich pale green natural silicate
Synthetic product ~CuHAsO3 (18th century)
Copperacetoarsenite, a pigment first produced around 1800 with the formula Cu(CH3COO)23 Cu(AsO2)2. Later “improvements”
Mineral 2CuCO3.Cu(OH)2
Pigment from the mineral Lapis lazuli, with the approximate composition (Na,Ca)8(AlSiO4)6(S,SO4,Cl)
Table 2. Pigments found on the examined medieval objects, with some chemical and mineralogical data.
22 Anders G. Nord et al.
Church Hejde Hogrän Endre Etelhem Stånga Gerum Barlingbo Burs
Workshop B B H H H M MB E
Lead white X X X X X X
Plattnerite X X X X X
Massicot X X?
Ochre X
Orpigment X X
Gold X X X
Iron oxide X X X X X
Minium X X X X X X
Cinnabar X X
Caput mortuum X
Malachite X X
Atacamite X
Green earth X
Azurite X X X
Ultramarine X
Table 3. Summary of pigments detected on seven fonts and a limestone pew. Abbreviations for the workshops as in tab. 1. Lime is not included in the table. Nor are the copper arsenite “improvements” on the Burs pew.
Organic binding media
We analysed six samples with a Perkin Elmer Spectrum100 FTIR spectrometer at the Royal Institute of Technology in Stockholm, aiming to search for organic binding media. Another larger sample from the pew we sent to the Institut Roy
al du Patrimoine Artistique (IRPA) in Brussels for GCMS analysis (PolarisQ and TripelTOF 5600 instruments). These attempts to trace organic binding media, e.g. proteinous glue, lin
seed oil or wax, all known to have been used on medieval artefacts, were not fruitful. The FTIR analyses only indicated the presence of calcium carbonate in addition to the pigments. The larg
er sample from the limestone pew, analyzed at IRPA, contained low concentrations of stearic and palmitic acid, as well as traces of the proteins keratin and trypsin. However, these organic sub
stances may be contaminations from later activi
ty. We conclude that lime was used as a binding medium, and that any additional organic sub
stances which may have been used have long
since deteriorated and are no longer possible to detect.
Lead isotope data and possible origins of the pigments
We are interested in the origin of these medieval pigments. Common ones like green earth and iron oxides were most likely collected locally. All other pigments listed in tab. 3, including gold, must have been imported. Primarily this import involved countries in central and south Europe.
Since Visby was an important commercial centre during the Middle Ages, most pigments could probably be bought from merchants there. Obvi
ously even ultramarine (from the blue mineral lapis lazuli), which was notably only found in Afghanistan at that time, was available.
As regards lead, its stable isotopes have long been used by archaeologists to reveal the origin of the metal. The stable isotopes
206Pb,
207Pb and
208Pb form by decay of uranium and thorium, whereas
204Pb is a stable, nonradiogenic isotope.
(Their relative isotopic abundance is usually giv
en by various ratios. For instance
, 206Pb/
204Pb,
207Pb/
204Pb and
208Pb/
204Pb are used in geolo
gical literature, while
207Pb/
206Pb and
208Pb/
206
Pb are more commonly seen in archeological
applications). The decay is extremely slow, and
Pigment traces on medieval stonework in Gotland’s churches 23
Sample Worksh Pigment 206Pb/204Pb 207Pb/204Pb 208Pb/204Pb 207Pb/206Pb 208Pb/206PbHejde13 B Plattnerite 18.458 15.617 38.375 0.8460 2.0794
Hejde14 B Plattnerite 18.453 15.620 38.410 0.8464 2.0815
Hogrän6 B Plattnerite 18.456 15.621 38.368 0.8464 2.0795
Hogrän7 B Minium 18.407 15.612 38.329 0.8481 2.0828
Etelhem16 H Minium 18.473 15.623 38.397 0.8456 2.0788
Stånga35 H Minium 18.492 15.634 38.424 0.8453 2.0779
Burs25 E Plattnerite 18.347 15.625 38.201 0.8559 2.0929
Table 4. Lead isotope data obtained for seven lead pigments. Abbreviations for the workshops as in tab. 1.
Most samples were run in duplicate and analytical errors (2 σ of the mean) are ±0.10%, or better, for ratios involving 204Pb, and c. ±0.04% for the remaining ratios.
so the relative isotopic abundances depend on the geological age of the ore and the conditions under which it was mineralised. This forms the basis for using lead isotope data as a fingerprint for the origin of lead. Many studies have been undertaken on archaeological metal artefacts with lead as an alloying element, such as bronze objects and silver coins. A few studies of lead pig
ments have also been published (e.g. Brill et al.
1997; Fortunato et al. 2005; Nord et al. 2015).
In the present study, seven lead pigment sam
ples large enough for an isotope study were ana
lysed at the Swedish Museum of Natural History with a MCICPMS spectrometer (Microsoft Iso
probe). These samples contained minium (Pb3O4) and/or its oxidized transition form plattnerite.
They were analysed following standard proce
dures according to Ling et al. (2013). The results are summarised in tab. 4. The values for the seven samples are similar.
The data show that none of these lead pigments originate from ore in the Bergslagen region of Sweden (the principal medieval mining district, in the middle of Sweden). This ore district is geo
logically very old, around 1900–1800 million years (Ma), and its lead ores display a narrow range of isotope ratios, with
206Pb/
204Pb clustering around 15.70. This is quite different from the analysed samples with ratios in the range 18.34–18.49. Swe
dish lead was thus certainly not used for the pig
ments identified in the present study. Instead the
isotope data indicate origins in much younger Palaeozoic to Mesozoic (545–65 Ma) deposits.
Using this information to determine the ori
gins of the lead is difficult. Lead isotope data have not been published for all possible ore districts.
Also the isotopic fingerprints of some ore dis
tricts overlap. The data are graphed in fig. 2. The samples Hogrän7 and, particularly, Burs25 with
206Pb/
204Pb=18.347, are slight exceptions, whilst the other five samples have uniform composi
tions with
206Pb/
204Pb clustering around 18.46.
The small deviations among the isotope ratios indicate that all probably originate from a re
stricted area. This is consistent with the fact that all seven baptismal fonts were made during the 12th century, involving contemporary sculptors, who bought pigments from merchants trading painting materials from only a few ore districts.
In a previous lead isotope study (Nord et al.
2015), 28 pigment samples from medieval murals (c. 1150–1500) in Swedish churches, seven of them on Gotland, were analysed. The provenancing indicated Germany as the most likely origin. This tentative conclusion was based on the following facts. 1) With few exceptions, the lead isotope data of the 28 samples matched well with data published for ore mined in the Harz and Erzge
birge regions (Niederschlag et al. 2003). 2) The nearest large ore districts were in Germany. 3) Secondary minerals found in malachite inde
pendently indicated German ores (Nord et al.
2012). 4) Many medieval painters in Sweden were Germans or influenced by German art.
Thus it seems that in 1150–1500 lead pigments used in Sweden most likely originated mainly from Germany.
The present study’s results also point towards German ore districts. A number of criteria, such as mining history, metal content and lead isotope data, are necessary to identify the most likely mining districts for the seven samples. A candi
date must of course have been mined during the 12th century, and mined mainly for lead ore.
According to Niederschlag et al. (2003), mining
in Germany at this time was restricted to only a few areas. Based on published mineralogical data (Wittern 2001) and lead isotope data (Lévêque
& Haak 1993; Niederschlag et al. 2003; Monn et al. 2008), we conclude that the Harz and Erzge
birge are the most likely origins for the lead pig
ments, but the nearby Eisleben district is also a likely candidate (cf. fig. 3). The isotope match with these ores is good (cf. fig. 2). Note that the present isotope data for the Burs pew, made by Egypticus, are similar to two lead pigments from murals in Lye Church, also painted by Egypticus (Nord et al. 2015).
24 Anders G. Nord et al.
Fig. 2. Lead isotope data plotted as 208Pb/206Pb versus 207Pb/206Pb for lead pigments from the baptismal fonts (circles) and the limestone pew (bold circle). Reference data (Niederschlag et al. 2003) from various ore districts in Germany are included.
(Eis=Eisleben, Erz=Erzgebirge).
Fig. 3. Map of northern Europe indicating some possible sources of pigments. E=Eisleben, K=Kams
dorf (a less likely source for the pigments). V=Visby, B=Berlin.
Pigment traces on medieval stonework in Gotland’s churches 25 Conclusions
None of our lead pigment samples originates from a Swedish mine. The lead mines of Bergslagen were worked already from AD 900, but during the Middle Ages the miners probably did not know how to prepare lead pigments, or did not find it commercially profitable. With current knowledge, the most likely conclusion seems to be that the lead originates from Germany, prob
ably from the Harz and Erzgebirge but possibly also from Eisleben. This is also in agreement with earlier results (Nord et al. 2012; 2015) and the fact that many Swedish churches were painted by Germans, or in a German style.
As regards the other pigments identified on the examined objects, common pigments were probably obtained locally. Azurite and malachite are rare in Swedish bedrock and must have been imported. During the Middle Ages, cinnabar was mainly mined in Spain and the western Balkans.
The merchants of Visby, an important member
town of the Hanseatic League, provided cinnabar, orpigment, ultramarine and gold from southern Europe or Asia. Or these materials were brought to Gotland by foreign painters. Thus while the sandstone for baptismal fonts was quarried locally, the pigments for their decoration have quite a diverse background. This shows that trade between Gotland and the European continent was well developed already around AD 1150.
Acknowledgements
Thanks to the Berit Wallenberg Foundation for a generous research grant (BWS2013.0010), to the staff of the Swedish Museum of Natural His
tory (Stockholm) for valuable help with the ana
lytical instruments and comments on our manu
script, and to Marina van Bos (IRPA, Brussels) for her analytical contribution with the Burs sample. Thanks finally to all the parishes involv
ed for their kind help with sampling and docu
mentation.
References
Aze, S., Valler, J.M., Pomey, M., Baronnet, A. & Grauby, O., 2007. Red lead darkening in wall paintings:
natural ageing of experimental wall paintings versus artificial ageing tests. European Journal of Mineralogy 19. Stuttgart.
Aze, S., Valler, J.M., Detalle, V., Grauby, O. & Baron
net, A., 2008. Chromatic alterations of red lead pigments in artwork: a review. Phase Transitions 81.
Yverdon.
Brill, R.H., FelkerDennis, C.C., Shirahata, H. & Joel, E.C., 1997. Lead isotope analysis of some Chinese and central Asian pigments. Agnew, N. (ed.). Con
servation of Ancient Sites on the Silk Road. Los Ange
les.
Fortunato, C., Ritter, A. & Fabian, B., 2005. Old mas
ters’ lead white pigments: investigations of paint
ings from the 16th and 17th centuries using high precision lead isotope abundance ratios. Analyst 130. London.
Giovannoni, S., Matteini, M. & Moles, A., 1990. Stu
dies and developments concerning the problem of altered lead pigments in wall paintings. Studies in Conservation 35. London.
Hauff, G., 1988. Study and conservation of the poly
chrome portal sculptures of the Holy Cross min
ster in Schwäbisch Gmünd. Ciabach, J. (ed.). VIth International Congress on Deterioration and Conserva
tion of Stone. Torun.
Kotulanová, E., Bezdička, P., Hradil, D., Hradilová, J., Svarcová, S. & Grygar, T., 2009. Degradation of leadbased pigments by salt solutions. Journal of Cultural Heritage 10. Amesterdam.
Lagerlöf, E., 1999. Gotland och Bysans. Bysantinskt infly
tande på den Gotländska kyrkokonsten under medelti
den. Visby.
Lévêque, J. & Haak, U., 1993. Pb isotopes of hydro thermal ores in the Harz. Monograph series on Mine
ral Deposits 30. Berlin.
Lindblom, A. & Svahnström, G., 1959. Gotländska sten
mästare. Malmö.
Ling, J., HjärtnerHoldar, E., Grandin, L., Billström, K. & Persson, P.O., 2013. Moving metals or indige
nous mining? Provenancing Scandinavian Bronze Age artefacts by lead isotopes and trace elements.
Journal of Archaeological Science 40. London.
Lundquist, S., 2012. Medeltida stenmästare och dopfuntar på Gotland. Romanska skedet, 1100–1200talen.Burgsvik.
Monn, F., Hamer, K., Lévêque, J. & Sauer, M., 2008.
Pb isotopes as a reliable marker of early mining and smelting in the Northern Harz province (Lower Saxony, Germany). Journal of Geochemical Explo
ration 68. Amsterdam.
Niederschlag, E., Pernicka, E., Seifert, Th. & Bartelheim, M., 2003. The determination of lead isotope ratios by multiple collector ICPMS: a case study of Early Bronze Age artefacts and their possible relation with ore deposits of Erzgebirge. Archaeometry 45.
Oxford.
Nord, A.G. & Tronner, K., 2000. Chemical analysis of mediaeval mural paintings in Sweden. Goupy, J. &
Mohen, J.P. (eds). Art et Chimie la Couleur, Actes du Congrès CNRS 1998. Paris.
26 Anders G. Nord et al.
– 2011. Kemisk analys av fjorton medeltidskyrkors muralmålningar. Fornvännen 106.
– 2014. Färganalys av fem gotlandskyrkors portaler och muralmålningar. Fornvännen 109.
Nord, A.G., Tronner, K. & Björling Olausson, K., 2012. Copper vanadate minerals found in mediae
val mural paintings. Studies in Conservation 57. Lon
don.
Nord, A.G., Tronner, K., Nisbeth, Å. & Göthberg, L., 1996. Färgundersökningar av senmedeltida kalk
måleri – Härkeberga, Täby, Härnevi och Risinge kyrkor. Konserveringstekniska Studier 12. Stockholm.
Nord, A.G., Billström, K., Tronner, K. & Björling Olausson, K., 2015. Lead isotope data for prove
nancing mediaeval pigments in Swedish mural paintings. Journal of Cultural Heritage 16. Amster
dam.
Petushkova, J.P. & Lyalikova, N.N., 1986. Microbio
logical degradation of leadcontaining pigments.
Studies in Conservation 31. London.
Roosval, J., 1918. Die Steinmeister Gottlands. Stockholm.
– 1925. Revision av gotländska dateringar. Fornvän
nen 20.
RossiManaresi, R., 1981. The polychromy of the 13th century stone sculpture in the façade of the Ferrara Cathedral. ICOM Committee for Conservation, 6th triennial meeting, Ottawa, 21–25 September 1981: pre prints. Paris.
– 1984. The polychromy of the portals of the Gothic Cathedral of Bourges. ICOM 7th triennial meeting, Copenhagen, 1014 September 1984: preprints. Paris.
Vasilyeva, S., 2009. Bysantinska traditioner i Gotlands konst under 1100talet. Fornvännen 104.
Wittern, A., 2001. Mineralfundorte in Deutschland. Stutt gart.