Journal of Archaeological Science 36 (2009) 2493–2501

Contents lists available at ScienceDirect

Journal of Archaeological Science

journal homepage: http://www.elsevier.com/locate/jas

Classification and provenance of soapstones and garnet chlorite schist

artifacts from Medieval sites of Tuscany (Central Italy): insights into
the Tyrrhenian and Adriatic trade
Patrizia Santi a, *, Alberto Renzulli a, Fabrizio Antonelli b, Antonio Alberti c
a
Dipartimento di Scienze Geologiche, Tecnologie Chimiche e Ambientali, Università degli Studi di Urbino, ‘‘Carlo Bo’’, 61029 Urbino, Italy
b
Laboratorio di Analisi dei Materiali Antichi, LAMA, Università Iuav di Venezia, 30125 Venezia, Italy
c
Dipartimento di Scienze dell’Antichità e del Vicino Oriente, Università Cà Foscari di Venezia, 30123 Venezia, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Soapstones (talc-bearing schists) and garnet chlorite schist artifacts found in Medieval archaeological
Received 17 February 2009 sites of Tuscany (Central Italy) were classified, in order to define provenance of the different lithotypes. In
Received in revised form Italy and throughout the Central Europe, these greenschist facies metamorphic rocks are generally
28 April 2009
known, among the archaeologists, as the pietra ollare from the Alps. The investigated Tuscan archaeo-
Accepted 17 May 2009
logical sites are between 6 and 13th century AD and were strictly linked, in that period, to the well
defined network trade running along Tyrrhenian coast. Samples come from little containers used for
Keywords:
cooking and preserving food and showing traces of lathe manufacturing at their sidewalls. According to
Metamorphic rocks
Soapstones modal mineralogy, petrographic texture, XRD, SEM-EDS and whole rock chemistry we recognised, among
Schists the 18 studied findings, three different petrographic groups of the Alpine pietra ollare. (i) Fine grained
Alpine pietra ollare magnesite talc schists (i.e. soapstones) from outcrops of the Central Alps located in the Valchiavenna
Archaeometry area. (ii) Garnet chlorite schists from the Valle d’Aosta region. (iii) Amphibole talc schists (i.e. soapstones)
Middle age with a provenance in the Ticino area. It is worth noting that artifacts of pietra ollare lithotypes from the
Tuscany Western Alps (i.e. garnet chlorite schists and amphibole talc schists) were not detected in the archae-
ological sites of the Middle Adriatic coast of the Central Italy, belonging to the same Medieval time
interval. This emphasises that the petrographic groups of pietra ollare from the Alps spread to the south
of the Po Plain according to Western and Eastern trade along the Italian Peninsula, using respectively, the
Tyrrhenian and the Adriatic Sea commercial routes.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction
The Italian term pietra ollare (from the Latin olla, i.e. container) is
normally used by the European archaeologists for artifacts of basic
and ultrabasic metamorphic rocks, mostly belonging to the
greenschist facies. All of the rocks referred to as pietra ollare are
geologically linked to the Apennine-Alpine ophiolite axis which
spread out for a long distance from Liguria (Italy) to the Stiria
(Austria; Dietrich, 1980). These metamorphic rocks were highly
prized for the production of pots and food containers because well
workable (related to the presence of talc) and their resistance to fire
(Antonelli et al., 2006). One of the most common lithotypes of
Alpine pietra ollare is grey to pale-green, coarse to fine grained
rocks consisting of talc, chlorite and carbonate minerals (mostly
* Corresponding author.
E-mail address: patrizia.santi@uniurb.it (P. Santi).
magnesite) cropping out in the Central and Western Alps (Mannoni
et al., 1987). These magnesite-bearing talc schists are soapstones,
usually derived from carbonate-free ultramafic rocks by interaction
with an externally derived CO2–H2O fluid phase (Bucher and Frey,
2002). In fact, serpentinites represent the usual hydrated low-
temperature alteration of mantle peridotites (lherzolites or harz-
burgites), but small amounts of CO2 in the fluids are sufficient to
convert serpentine assemblages into carbonate-bearing rocks.
According to Bucher and Frey (2002) the three major types of
carbonate-bearing ultramafic rocks are: ophimagnesites
(antigorite þ magnesite), soapstones (talc þ magnesite) and sag-
vandites (enstatite þ magnesite). The term soapstone (synony-
mous: steatite) is however commonly used for any talc-bearing
schist. Pietra ollare groups soft grey-coloured (soapstones) and
harder green-coloured lithotypes (chlorite schists, serpentine
schists and amphibole schists) both very widespread among the
artifacts of the archaeological sites throughout Italy and Central
Europe. A useful mineralogical data-base of the different

0305-4403/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jas.2009.05.006
2494 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501
metamorphic rocks known as pietra ollare and cropping out in the
Alps, is represented by the 11 petrographic groups (Table 1)
recognised by Mannoni et al. (1987). The Central Alps quarries of
Valchiavenna and Valmalenco (magnesite talc schists and chlorite
schists), those of the Valle d’Aosta (garnet chlorite schists) and the
Ticino Valley (various types of schists) in the Western Alps, were
the most exploited sources in the whole Medieval period (Table 1;
Mannoni et al., 1987). The use of the green-coloured pietra ollare is
documented from the pre-Roman Period (Mannoni et al., 1987). The
grey-coloured soapstones coming from Valchiavenna started to be
used in the 1st century AD (Bolla, 1991). Valle d’Aosta
green lithologies (garnet chlorite schists) became very widespread
starting in the 4th century AD (Mollo Mezzena, 1987) with artifacts
largely distributed around the outcropping areas of the Western
Alps (Piemonte, Liguria and Provence). The use of the green-col-
oured pietra ollare gradually decreased with time in the Middle Age
whereas the use of grey-coloured rocks (i.e. soapstones) increased
(Lusuardi Siena and Sannazaro, 1987). This seems to be confirmed
by recent findings of pietra ollare from Medieval archaeological
strata of additional excavations in Pisa (works are still in progress)
where the proportion of grey-coloured lithotypes strongly increase
with time.
Provenance of pietra ollare from Alpine quarries and production
centres was established for numerous findings in Roman and
Middle Age archaeological sites of (i) Northern Italy (Mannoni and
Messiga, 1980; Mannoni, 1986; Mannoni et al., 1987; Alberti, 1997;
Bonazza et al., 1999; Malaguti and Zane, 1999) and (ii) Adriatic
regions of the Central-Southern Italy (Mannoni, 1986; Alberti, 1997;
Santi et al., 2005). No comparable detailed mineralogical, petro-
graphic and geochemical data exist for the pietra ollare artifacts
found in Medieval (6–13th century AD) archaeological sites of
Tuscany. In this work we present a petrographic classification and
relative provenance of 18 pietra ollare fragments coming from
archaeological excavations located in the provinces of Pisa and
Livorno and the cities of Pisa and Florence (Fig. 1). Although the
most useful parameter to unravel source areas of the pietra ollare
groups is the modal mineralogy, we also compared the whole rock
chemical data of the Tuscan findings with the available literature
data of the Alpine outcrops such as those of the Valle d’Aosta
(Mugnaini, 1995), Valmalenco (Bonazza et al., 1999), and Valchia-
venna (Santi et al., 2005). Due to the chemical heterogeneity of the
single quarries, statistical processing of the whole rock chemical
data was not carried out. Whole rock chemical comparisons will be
used to support a provenance however well constrained by modal
mineralogy and texture. These new mineralogical, petrographic
Table 1
Summary on the macroscopic colour, modal mineralogy, classification, petrographic groups and provenance of the different pietra ollare lithotypes from the Alps
(according to Mannoni et al., 1987).
Macroscopic Modal
colour mineralogy
Light grey-green Talc, carbonates (magnesite),
chlorite, opaques
Light grey-green Talc, carbonates, chlorite,
amphibole, opaques
Green Chlorite, epidote, garnet,
chloritoid, opaques  talc
Light green Serpentine, chlorite, talc,
opaques  carbonates  olivine
Dark green Pyroxene, amphibole, talc  olivine
 mica  plagioclase
Dark green-grey Amphibole, chlorite, opaques  talc
 mica  quartz  epidote
Grey-white Olivine, talc, chlorite, opaques
 carbonates  serpentine
Green Chlorite, epidote, opaques +/-; albite  mica
 quartz
and whole rock chemical data will be finally compared with the
available provenances of the pietra ollare artifacts from other
archaeological sites of Central Italy, in particular of those con-
cerning Medieval findings of the same time interval found in the
regions of the Middle Adriatic coast (Marche and Abruzzo; Fig. 1).
A framework of the pietra ollare trade networks during the Middle
Age for the Central Italian regions will be finally outlined.
2. Exploitation technology and trade of the pietra
ollare in antiquity
The Alpine pietra ollare lithotypes were exploited since the
prehistoric age and widely used from the Roman Empire to the
entire Middle Age when the artifacts became very widespread
throughout Italy and Central Europe (Mannoni and Messiga, 1980;
Mannoni et al., 1987). The most ancient recorded artifacts as oil
lamps and crucibles, dating back to the Iron Age, come from some
valleys of the Central Alps (Mannoni and Messiga, 1980) in areas
located very close to the outcrops. The production of small artifacts
(containers for cooking and preserving food) with a local network
distribution, continued until the 1st century AD. Manufacturing of
the pietra ollare using the lathe is known since the Iron Age. The
introduction of big hydraulic lathes in the 4th century AD could
have produced enough pietra ollare to reach areas outside the
production zones (extra-regional trade). According to historic
documents (e.g. Naturalis Historiae, Pliny the Elder, 1st century AD;
Enciclopédie, Diderot, 1782) the use of pietra ollare for making
artifacts in the Alpine valleys persisted for almost 18 centuries. The
production of the pietra ollare containers attained a peak during the
5–7th century AD as shown by findings in archaeological sites of
Italy, Switzerland, France, Austria and Germany (Mannoni and
Messiga, 1980).
Two type of excavations for pietra ollare were generally recog-
nised, correlated with the nature of raw material outcrops:
(i) surface exploitation working erratic blocks transported along
fluvial or glacial valleys, far from the primary outcrops; (ii) opencast
quarries and underground galleries developed in lenses or veins
(Donati, 1986a). The exploitation of the pietra ollare in the Alps
removed entire outcrops leaving relatively few traces behind.
Numerous traces of ‘‘drillings’’ tests are represented by 20–25 cm
long (the average height necessary to make an artefact using the
lathe) holes which were made on the surface of the pietra ollare
outcrops to detect the rock suitability (Donati, 1986a). In some
Alpine artifacts, Mannoni (1986 and reference therein) and Donati
(1986b) recognised four different production techniques on the
Classification Petrographic groups Alpine Provenance
Magnesite talc C (coarse grained) Valtellina, Valbregaglia,
schists D (fine grained) Valchiavenna
Amphibole talc B (coarse grained) Ticino and Toce Valleys
schists E (fine grained)
Chlorite schists F (fine grained) Valle d’Aosta, Valtellina,
G (coarse grained) Valmalenco, Val di Lanzo (F)
Serpentine schists A Ticino and Toce Valleys
Meta-gabbros H Ticino and Toce Valleys (H)
Amphibole schists I Ticino Valley
Olivine schists K Ticino Valley
Prasinites L Rare outcrops
 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501 2495

Fig. 1. Pietra ollare findings from Medieval archaeological sites of Central Italy. Location of the Tuscan 6–13th century AD archaeological sites from the present work. 1: Pisa; 2:
Vicopisano (PI); 3: San Miniato (PI); 4: Florence; 5: Collesalvetti (LI); 6: Monteverdi (PI); 7: Populonia (LI); 8: Piombino (LI).

basis of the stylistic and technological analyses: (i) carving with
metallic tools, (ii) surface finishing by manual lathes (since the 1st
century AD), (iii) using horizontal manual and/or hydraulic lathes
(from 3 to 6th century AD); (iv) making several pots starting from
a single rock block (‘‘onion technique’’) using hydraulic horizontal
lathes (since the 17th century AD).
A standardisation of the pietra ollare artifacts, characterised by
the progressive thinning of the walls, was developed since the 4th
century AD as the result of using heavy lathes (Donati, 1986b;
Alberti, 1997). Most of the main production centres of pietra ollare
containers were located close to the quarries or in the nearest
areas located along the main waterways, rivers and/or lakes

Table 2
Archaeological sites of the pietra ollare samples from the present work. Modal mineralogy, petrographic classification and possible lithotypes of the Alpine
pietra ollare according to the groups of Mannoni et al. (1987) are also given. Abbreviations: Mgs ¼ magnesite, Tlc ¼ Talc, Chl ¼ Chlorite, Amph ¼ Amphibole,
Opq ¼ Opaque minerals, Grt ¼ Garnet.

Samples Location of the archaeological sites Modal mineralogy Petrographic Mannoni et al.
classification (1987)
Province of PISA
SMV ’98 US2038 Monastero di S.Michele alla Verruca, Vicopisano Tlc, Chl, Amph, Opq,  Mgs Amphibole talc schist E
(11th cen.)
PI-VF ’01 US1471 Via Facchini, Pisa (11th cen.) Tlc, Chl, Amph, Opq,  Mgs Amphibole talc schist E
PIPD I 173 Piazza Dante, Pisa (middle 10–middle 13th cen.) Tlc, Chl, Amph, Opq,  Mgs Amphibole talc schist E
PIPD II 19 (scattered) Piazza Dante, Pisa Tlc, Chl, Amph, Opq,  Mgs Amphibole talc schist E
PM 92 85 Piazza dei Miracoli, Pisa (6th cen.) Chl, Opq, Grt Garnet chlorite schist F–G
MV 2004 US1023 Monastero di S.Pietro di Monteverdi Tlc, Mgs, Chl, Opq Magnesite talc schist D
SGO1 (scattered) San Genesio, San Miniato Tlc, Mgs, Chl, Opq Magnesite talc schist D
SGO2 (scattered) San Genesio, San Miniato Tlc, Mgs, Chl, Opq Magnesite talc schist D
SGO3 US 1001 San Genesio, San Miniato Tlc, Chl, Amph, Opq,  Mgs Amphibole talc schist E
SGO4 ‘‘area 22 000’’ San Genesio, San Miniato Tlc, Mgs, Chl, Opq Magnesite talc schist D

Province of LIVORNO
PB ’00 S.1, UT1, S.4
SQ 2005 US1179
PCS 02, Sg.XX, US329
PCS 05, Sg.XX, US480
PCS 06, Sg.XXXIVb
PCS (scattered)
Collesalvetti (11–12th cen.) Tlc, Mgs, Chl, Opq Magnesite talc schist D
Monastero di S.Quirico, Piombino Chl, Opq, Grt Garnet chlorite schist F–G
Populonia (11–13th cen.) Tlc, Mgs, Chl, Opq Magnesite talc schist D
Populonia (11–13th cen.) Tlc, Mgs, Chl, Opq Magnesite talc schist D
Populonia (probably 11–13th cen.) Chl, Opq, Grt Garnet chlorite schist F–G
Populonia (11–13th cen.) Chl, Opq, Grt Garnet chlorite schist F–G

FLORENCE
PO1 phase 12 Via De’Castellani, Firenze (8–10th cen.) Chl, Opq, Grt Garnet chlorite schist F–G
PO4 phase 19 Via De’Castellani, Firenze (13th cen.) Tlc, Mgs, Chl, Opq Magnesite talc schist D
2496 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501
Fig. 2. Representative fragment of the pietra ollare studied artifacts, showing the
features of the external (a) and internal (b) walls.
(Gaggi and Leoni, 1984). Stones were carved near the quarries
because of the difficulty in the transport of big blocks due to their
weight (Donati, 1986a): the average density of the pietra ollare is
between 2500 and 3300 kg/m3 depending on the different litotypes
(Robertson, 1982; Antonelli et al., 2006).
3. Sampling and analytical methods
The studied fragments of pietra ollare (18 samples) come from
different Medieval archaeological sites located in Tuscany; in
particular, 10 samples from the province of Pisa, 6 from the prov-
ince of Livorno and 2 from Florence (Fig. 1 and Table 2). Most of
them come from little containers used for cooking and preserving
foods as emphasised by the smoky/burned external portion
(Fig. 2a). Thickness varies between 6 and 14 mm. Traces of lathe
Fig. 3. Representative thin section petrography and texture of some pietra ollare samples: (a) magnesite talc schist, (b) amphibole talc schist, (c) and (d) garnet chlorite schists.
Abbreviations: Mgs ¼ magnesite, Tlc ¼ Talc, Chl ¼ Chlorite, Amph ¼ Amphibole, Grt ¼ Garnet.
manufacturing are frequently present both in the external and
internal sidewalls (Fig. 2a, b).
Modal mineralogy analyses were done by optical polarising
microscope (thin sections) and X-Ray Diffraction (XRD) techniques.
XRD data come from a Philips PW1840 diffractometer (radiation
Cu Ka/Ni with 40 Kv and 20 mA; detection limit 5%; University IUAV
of Venice) using powdered samples. Scanning Electron Microscope
(SEM) qualitative analyses of mineral phases were carried out using
a Philips 515 equipped with a PV9900 Energy Dispersive System
(EDS) analyser (University of Urbino). Whole rock major (Induc-
tively Coupled Plasma-Optical Emission Spectrometry) and trace
(Inductively Coupled Plasma-Mass Spectrometry) elements anal-
yses were determined at the Activation Laboratories LTD (Ancaster,
Canada) using a ‘‘Thermo Jarrell-Ash ENVIRO II ICP’’ and a ‘‘Perkin
Elmer SCIEX ELAN 6000 ICP-MS’’ respectively. Whole rock analyses
errors are  1% for major oxides and 3% for trace elements.
4. Results: mineralogy, petrography and geochemistry
Petrography (Fig. 3), XRD (Table 3), mineral analyses (Fig. 4) and
whole rock chemical compositions (Table 4) recognised three
different lithotypes of pietra ollare among the studied samples.
4.1. Magnesite talc schists (soapstones)
Magnesite talc schists samples (Fig. 3a) are the most abundant
among the studied pietra ollare artifacts (44% of the total) and
correspond to the ‘‘D group’’ of Mannoni et al. (1987). They are
characterised by the typical mineralogy (see modal mineralogy in
Table 2) and texture of soapstones s.s.: fine to medium grained
rocks with idioblastic talc, heteroblastic magnesite and clinochlore
(Fig. 4). Rare opaques minerals (ilmenite, magnetite and pyrite) are
also present. The sample coming from ‘‘Collesalvetti’’ (Livorno
province; Table 2) contains a great amount of magnesite with a well
 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501 2497

Table 3
XRD analyses of the pietra ollare samples. þþþ ¼ very abundant, þþ ¼ average abundance, þ ¼ present,  ¼ traces.

Magnesite talc schists

Sample MV 2004 SG01 SG02 SG04 PB ’00 PCS 02 PCS 05 PO4

Talc þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ
Clinochlore þþ þþ þ þþþ þ þþþ þþ þþ
Magnesite þ þþ þþ þþ þþ þþþ þþþ 
Dolomite þ þ þþ þ þ
Brucite 
Magnetite  þ

Amphibole talc schists Garnet chlorite schists

Sample SMV ’98 PI-VF ’01 PIPD I 173 PIPD II 19 SG03 PM 92 85 SQ 2005 PCS 06 PCS PO1
Talc þþþ þþþ þþþ þþþ þþþ þþþ
Clinochlore þþ þþ þþ þ þþ þþþ þþþ þþþ þþþ
Magnesite  
Dolomite þ
Calcite þ
Amphibole  þ þ þ þ þ
Brucite 
Magnetite    

3500 5000

Counts
Si Clinochlore 4500 Si Talc
3000
Mg 4000
Sample: SMV '98 Sample: PO4
2500 3500
Counts

2000 3000
2500
1500 Al Mg
2000

1000 1500
1000
500 Fe
Cr Kev 500
Fe Kev
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8

4000 2500
Almandine Actinolite
Counts
Counts

3500
Si 2000 Si Sample: PO1
3000 Sample: PO1

2500
1500
2000
Al
1500 1000
Fe
1000
500 Al
Ca Fe
500 Ca Mg
Kev Kev
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8

4500 3000
Chloritoid Epidote
Counts

Si
Counts

4000 Al
Sample: SQ 2005 2500 Sample: SQ 2005
3500

3000 2000
Si Al
2500
1500
2000
Ca
1500 1000
Fe
1000
Mg 500 Fe
500 Kev Cr Kev
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8

Fig. 4. Representative EDS spectra of some minerals in the studied samples.

 2498
Table 4
Major and trace elements analyses of the pietra ollare artifacts (ICP-OES-MS). PI ¼ samples from Pisa or province of Pisa; LI ¼ samples from the province of Livorno; FI ¼ samples from Florence.

Magnesite talc schists Amphibole talc schists Garnet chlorite schists

PI PI PI PI LI LI LI FI PI PI PI PI PI PI LI LI LI FI

MV 2004 SG01 SG02 SG04 PB ’00 PCS 02 PCS 05 PO4 SMV’98 PI-VF ’01 PIPD I 173 PIPD II 19 SG03 PM 92 85 SQ 2005 PCS 06 PCS PO1
SiO2 44.50 35.43 33.83 34.28 36.08 35.98 37.49 48.35 53.54 50.72 47.76 49.59 51.55 25.19 24.30 23.83 24.40 28.33
Al2O3 2.32 2.07 0.74 2.04 2.20 2.70 2.06 2.43 3.23 3.69 5.73 1.95 1.13 19.94 21.66 20.28 24.17 16.81
TiO2 0.04 0.06 0.01 0.08 0.10 0.05 0.05 0.08 0.10 0.12 0.06 0.07 0.14 0.43 0.16 0.35 0.40 0.52
Fe2O3 (t) 5.51 7.01 6.37 6.71 7.48 7.42 7.11 7.18 6.11 12.02 11.29 13.72 15.05 35.79 34.49 36.68 33.83 34.70
MnO 0.09 0.13 0.10 0.12 0.12 0.12 0.11 0.06 0.09 0.12 0.20 0.19 0.16 0.20 0.04 0.05 0.06 0.15
MgO 29.63 34.03 33.94 33.12 32.48 34.06 33.28 27.68 30.63 27.92 27.61 25.07 26.60 10.41 11.50 10.35 9.93 9.85
CaO 3.53 1.15 0.37 1.48 1.62 1.17 1.24 3.41 0.23 0.17 0.49 3.31 0.71 0.42 0.08 0.03 0.16 1.08
Na2O 0.05 0.02 0.02 0.06 0.16 0.03 0.19 0.04 0.10 0.05 0.06 0.13 0.24 0.03 0.03 0.01 0.05 0.51
K2O <0.01 <0.01 <0.01 <0.01 <0.01 0.02 <0.01 <0.01 0.03 0.02 <0.01 0.05 0.06 <0.01 <0.01 <0.01 0.03 0.03

P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501

P2O5 0.02 0.04 0.05 0.01 0.08 0.11 0.12 0.05 0.11 0.02 0.08 0.18 0.03 0.03 <0.01 0.02 0.06 0.12
LOI 13.17 21.07 23.09 20.33 18.23 18.69 19.24 10.32 5.99 4.95 6.33 4.76 4.12 6.95 7.98 7.94 7.58 6.67

Total 98.79 101.00 98.54 98.21 98.55 100.40 100.90 99.58 100.20 99.79 99.62 99.03 99.79 99.37 100.20 99.49 100.70 98.78

Sc 13 9 7 9 10 12 12 8 10 4 5 5 5 26 23 29 16 41
V 40 37 26 29 41 44 43 69 40 21 11 17 43 95 87 136 84 233
Ba 11 8 9 10 20 21 19 19 29 4 9 16 12 9 4 4 4 47
Sr 19 9 4 11 13 8 9 30 6 <2 6 61 14 3 <2 <2 8 15
Y 6 <2 <2 <2 2 <2 <2 <2 6 <2 <2 <2 <2 62 37 84 23 47
Zr 5 5 <4 <4 10 <4 4 5 <4 10 7 4 10 393 100 318 494 39
Cr 1900 1550 1620 1590 1670 1990 1490 1860 3760 2070 4240 2990 3090 2450 1450 1240 780 <20
Co 68 87 70 82 80 61 80 61 75 125 134 128 52 249 353 250 216 271
Ni 1220 1540 1250 1240 880 1240 1420 980 1040 960 1390 1160 710 420 300 270 230 140
Cu <10 20 <10 20 40 90 30 90 50 70 50 70 280 <10 <10 <10 20 160
Zn 50 40 30 40 60 50 40 50 100 110 140 150 100 60 130 40 40 70
Ga 3 2 <1 2 3 3 2 4 10 7 9 5 4 25 25 21 23 29
Rb <2 <2 <2 <2 <2 <2 <2 <2 <2 <2 2 <2 <2 <2 <2 <2 <2 <2
Nb <1 <1 <1 <1 <1 <1 <1 <1 11 1 <1 <1 1 3 3 4 4 2
La 0.3 <0.1 0.2 0.1 0.4 0.2 0.1 0.1 0.5 0.1 0.2 0.6 3.9 3.6 0.9 15.9 8.8 8.4
Ce 0.8 <0.1 0.3 0.2 0.9 0.4 0.3 0.4 1.1 0.2 0.4 1.8 7.6 10.2 2.1 50.9 25.2 22.2
Pr 0.16 <0.05 <0.05 <0.05 0.14 0.05 <0.05 0.07 0.18 <0.05 0.06 0.28 0.76 1.64 0.35 7.27 3.55 3.16
Nd 1 0.1 0.2 0.2 0.8 0.3 0.2 0.6 1.1 0.2 0.3 1.6 2.6 8.9 2.3 37.4 19 16.5
Sm 0.4 <0.1 <0.1 <0.1 0.2 0.1 <0.1 0.3 0.4 <0.1 <0.1 0.5 0.5 2.8 0.9 11.4 6 5.4
Eu 0.14 <0.05 <0.05 <0.05 0.06 <0.05 <0.05 0.06 <0.05 <0.05 <0.05 0.08 0.05 0.2 <0.05 0.18 0.21 0.33
Gd 0.8 0.1 <0.1 0.1 0.3 0.2 0.1 0.5 0.7 <0.1 <0.1 0.5 0.4 7.5 3.1 15.7 7.6 7.8
Tb 0.2 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.2 <0.1 <0.1 <0.1 <0.1 1.6 0.8 2.9 1 1.5
Dy 1.1 0.2 <0.1 0.2 0.4 0.3 0.2 0.5 1.2 <0.1 <0.1 0.5 0.3 10.5 5.8 16.9 5 9
Ho 0.2 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.3 <0.1 <0.1 <0.1 <0.1 2.1 1.3 3.1 0.9 1.8
Er 0.7 0.2 <0.1 0.2 0.3 0.2 0.1 0.3 0.8 <0.1 <0.1 0.3 0.2 6.4 4.1 8.9 2.6 5
Tm 0.1 <0.05 <0.05 <0.05 0.05 <0.05 <0.05 <0.05 0.12 <0.05 <0.05 <0.05 <0.05 1.02 0.67 1.34 0.43 0.76
Yb 0.7 0.2 <0.1 0.2 0.4 0.2 0.1 0.2 0.8 <0.1 <0.1 0.3 0.2 6.8 4.3 8.2 2.9 4.8
Lu 0.09 <0.04 <0.04 <0.04 0.07 <0.04 <0.04 <0.04 0.1 <0.04 <0.04 <0.04 <0.04 1.02 0.64 1.11 0.44 0.64
Hf <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 0.2 <0.2 <0.2 0.2 9.3 3.4 10.2 12.8 1.4
Th <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.4 0.4 0.3 0.5 0.6 0.2 0.2 1.2 0.2 0.3
U 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 2.9 <0.1 <0.1 0.3 0.1 0.1 <0.1 0.5 0.3 0.2
 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501 2499

developed idioblastic crystals (up to 4 mm). The chemical compo-
sition of all the magnesite talc schists samples (Table 4) shows SiO2
33.8–48.3 wt%, low Al2O3 (0.7–2.7 wt%) and high MgO (27.7–
34.1 wt%) values (Fig. 5). Magnesite is also responsible for the very
high LOI values (10.3–23.1 wt%; Table 4) of this pietra ollare lith-
otype. Some other (undetected) accessory silicate minerals are
present in the two samples showing the highest SiO2 contents (MV
2004 and PO4; Table 4) which are also characterised by an elevated
CaO content of 3.4–3.5 wt% due to the presence of dolomite
(detected by XRD; Table 3). In all of the samples without dolomite,
CaO content is in fact restricted to 0.4–1.6 wt%. Magnesite talc
schists show elevated contents of Cr (1490–1990 ppm) and
Ni (880–1540 ppm). Very low LREE (e.g. Ce  0.9 ppm) and HREE
(e.g. Yb  0.7 ppm) contents seem to exclude the presence of
accessory phases in which REE can enter the mineral structures.
Rare brucite was detected by XRD analyses (Table 3).
4.2. Amphibole talc schists (soapstones)
Amphibole talc schists (corresponding to the ‘‘E group’’ of
Mannoni et al., 1987) represent 28% of the studied pietra ollare
fragments. They consist of fine grained rocks with heteroblastic talc
(Fig. 3b) and clinochlore as the major minerals and minor amount
of actinolitic amphibole (Fig. 4). Opaque minerals (magnetite and
pyrite) and magnesite can be present. This modal mineralogy (Table
2) is well reflected in the whole rock chemistry (Table 4) as these
samples show high SiO2 (47.8–53.5 wt%) and MgO (25.1–30.6 wt%)
and relatively low Al2O3 (1.1–5.7 wt%) values (Fig. 5). High Cr
contents (2070–4240 ppm) reflect the presence of chromian
magnetite, Ni (710–1390 ppm) and HREE (e.g. Yb  0.8 ppm)
contents are comparable to those of the magnesite talc schists. LREE
are relatively high (e.g. Ce up to 7.6 ppm) and this may be due to the
presence of some accessory phase (epidote?) although not detected
by XRD or optical microscope.
4.3. Garnet chlorite schists
Garnet chlorite schists correspond to the ‘‘F and G groups’’ of
Mannoni et al. (1987) and represent 28% of the total investigated
samples (see modal mineralogy in Table 2). This group consists of
medium to fine grained rocks with abundant heteroblastic green
clinochlore, opaque minerals and rounded almandine garnet (Figs.
3c, d and 4). Rare crystals of quartz were also detected. Chloritoid
and epidote are also present (Fig. 4), the former being characterised
by crystals up to 1.6 mm in size (sample SQ2005) and indigo to blue
pleocroism. Rare brucite is detected by XRD analyses (Table 3).
All the samples have a well developed schistosity, locally with a
decussate texture (Fig. 3d). They are chemically detectable (Table 4)
with high Al2O3 (16.8–24.2 wt%) as the result of abundant chlorite.
Fe2O3 tot (33.8–36.7 wt%) prevails over MgO (9.8–11.5 wt%) whereas
SiO2 (23.8–28.3 wt%) is the lowest among the three lithotypes.
Relatively high V (84–233 ppm) and Sc (16–41 ppm) with respect to
the other two petrographic groups, can be attributed to the
magnetite-spinel series. High Fe2O3 tot, HREE (Gd up to 15.7 ppm; Yb

44
25 Garnet chlorite
schists 39

20 34

15 29
Al2O3

MgO

Magnesite talc schists

24
10 Amphibole talc Amphibole talc
schists schists
19
5 Magnesite talc schists
14 Garnet chlorite
schists
0 9
22 27 32 37 42 47 52 22 27 32 37 42 47 52
SiO2 SiO2

40 7000
Garnet chlorite
35 schists 6000
30 Amphibole talc
5000
schists
25
4000
Fe2O3

Cr

20 Garnet chlorite
Amphibole talc schists 3000 schists
15
2000
10

5 1000 Magnesite talc schists

Magnesite talc schists
0 0
22 27 32 37 42 47 52 22 27 32 37 42 47 52
SiO2 SiO2

Findings from the province of Pisa Findings from the province of Livorno Findings from Florence

Valchiavenna outcrops (Santi et al. 2005) Valmalenco outcrops (Bonazza, 1999) Valle d'Aosta outcrops (Mugnaini, 1995)

Fig. 5. Binary diagrams for the pietra ollare studied artifacts compared with literature data of Alpine outcrops.
2500 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501
Fig. 6. The Western (Tyrrhenian Sea) and Eastern (Adriatic Sea) trade of the Alpine
pietra ollare during the Middle Age. Arrows indicate different provenance of the
petrographic groups. Filled circles show the pietra ollare findings in the middle Adriatic
(Santi et al., 2005) and Tyrrhenian (this work) regions. VA: Valle d’Aosta; TV: Ticino
Valley; VC: Valchiavenna.
up to 8.2 ppm; Lu up to 1.1 ppm), Hf (up to 12.8 ppm), Y (23–84 ppm)
and Zr (up to 494 ppm) values are well related to the modal
almandine crystals (Table 4). Relatively high contents of LREE (La up
to 15.9 and Ce up to 50.9 ppm) are linked to the presence of modal
epidote. The PO1 sample (Table 2) shows a slightly different modal
composition (Fig. 3d), with clinochlore, actinolitic amphibole,
magnetite, ilmenite, almandine garnet  albite. This distinctive
mineralogical assemblage is also reflected in the variation diagrams
(Fig. 5) where the PO1 sample shows the highest SiO2 (28.3%) and
the lowest Al2O3 (16.8 wt%) contents among the samples of this
petrographic group. The relatively high Ba (47 ppm) with respect to
the other samples of garnet chlorite schists (Ba 4–9 ppm) is certainly
due to the presence of some albite crystals (Na2O 0.5 wt% in the
whole rock chemistry).
5. Discussion
The archaeological sites in the Livorno and Pisa provinces and
Florence all contain magnesite talc schists and garnet chlorite
schists; amphibole talc schists are represented only by samples
from Pisa province (Table 2).
Soapstones, detected in all the sites, are the most abundant (13
total samples consisting of both magnesite talc schists and
amphibole talc schists) whereas garnet chlorite schists are subor-
dinate (5 samples). According to the archaeometric literature
(Mannoni et al., 1987) the different groups of Medieval pietra ollare
discovered in the Italian archaeological sites come from the Alps.
Grey-coloured pietra ollare lithotypes (Table 1) consisting of
talc þ chlorite þ magnesite come from the Central Alps
(Valchiavenna, Valtellina and Valbregaglia), whereas amphibole
talc schists have a provenance from the Western Alps (Ticino and
Toce valleys; Table 1; Donati, 1986b; Pfeifer and Serneels, 1986;
Mannoni et al., 1987). The most abundant green-coloured pietra
ollare lithotypes (Table 1), where chlorite is the prevailing mineral,
come from both the Western Alps (Valle d’Aosta) and the Central
Alps (Valtellina and Valmalenco) according to Pfeifer and Serneels
(1986) and Mannoni et al. (1987).
5.1. Provenance of soapstones
Among the investigated samples, the large presence (72%) of
fine grained, grey-coloured soapstones (‘‘D and E groups’’ of
Mannoni et al., 1987; Table 1) is probably due to the fact that these
rocks are well workable by means of lathe. Concerning all the
Medieval archaeological findings of pietra ollare studied in the
Middle Adriatic coast of the Central-Southern Italy (5–15th century
AD; Marche and Abruzzo regions; Santi et al., 2005) only soap-
stones belonging to the magnesite talc schist ‘‘D group’’ (Mannoni
et al., 1987), are present. Findings of pietra ollare from the Adriatic-
Ionian regions of Southern Italy (artifacts from Ugento, Lecce)
belong to the same group of magnesite talc schists (Mannoni, 1986).
Modal mineralogy of the magnesite talc schists from the
‘‘D group’’ of Mannoni et al. (1987) indicates that quarries are mainly
located in Valchiavenna (Central Alps). Compositions of the eight
Tuscan findings belonging to this group are clustered in and around
the range of the low Al2O3 and Fe2O3 and high MgO contents of the
Valchiavenna quarries (Fig. 5). We can summarise that the fine
grained magnesite talc schists found in all the archaeological sites of
Tuscany, come from the Central Alps, in particular from the well
known quarries located in the Valchiavenna area.
The five Tuscan findings made of fine grained amphibole talc
schists probably represent the ‘‘E group’’ (Table 1) of Mannoni et al.
(1987). This modal mineralogy suggests provenance from the
Ticino valley (Western Alps). Comparison with mineralogy and
petrography given by Pfeifer and Serneels (1986) for some outcrops
of amphibole talc rocks from the Ticino valley, seem to support this
provenance. Unfortunately, no chemical data are available for these
outcrops. The five samples of amphibole talc schists are charac-
terised by high SiO2 and Cr, low Al2O3 and intermediate MgO–Fe2O3
contents and do not match any quarries of Valchiavenna, Valma-
lenco and Valle d’Aosta (Fig. 5).
5.2. Provenance of garnet chlorite schists
The five artifacts made of garnet chlorite schists closely match
the ‘‘F group’’ of Mannoni et al. (1987; Table 1) from Valle d’Aosta
(Western Alps). The chlorite schists from Valmalenco are, however,
not so enriched in garnet. Only the Valle d’Aosta outcrops approach
the Fe2O3-rich, MgO-poor samples of the five Tuscan garnet chlo-
rite schists (Fig. 5).
6. Conclusions
The presence in the archaeological sites of Tuscany dating to
6–13th century AD of some pietra ollare lithotypes from Valle
d’Aosta and Ticino valley can be strictly associated with a Western
pietra ollare trade along the Tyrrhenian Sea (Alberti, 1997). The
circulation of these groups of pietra ollare towards the Middle
Adriatic coast was precluded by the presence of the Northern
Appenine barrier (Fig. 6). In the archaeological sites of 5–15th
century AD of the Middle Adriatic coast (Alberti, 1997; Santi et al.,
2005) only the soapstones coming from the Central Alps, were
found, supporting an Eastern pietra ollare trade, from the
 P. Santi et al. / Journal of Archaeological Science 36 (2009) 2493–2501
Valchiavenna area to the Adriatic Sea along the main waterways:
Mera River–Como Lake and finally the rivers Adda and Po. Never-
theless, the widespread talc magnesite schists from Valchiavenna
also reached Tuscany through the Western trade. We can conclude,
as shown in Fig. 6, that the existence of the Tyrrhenian and Adriatic
trade during the Middle Age determined the distribution of
different Alpine pietra ollare lithotypes in the regions of the Italian
peninsula to the south of the Po Plain.
Acknowledgements
We wish to thank C. D’Amico (University of Bologna) who
provided us many thin sections of pietra ollare samples from Valle
d’Aosta. We are also grateful to S. Swanson (University of Georgia)
and an anonymous reviewer for their useful suggestions.
References
Alberti, A., 1997. Produzione e commercializzazione della pietra ollare in Italia
settentrionale tra tardoantico e altomedioevo. In: 1 Congresso Nazionale di
Archeologia Medievale. Pisa, 1997. Proceedings, pp. 335–339.
Antonelli, F., Santi, P., Renzulli, A., Bonazza, A., 2006. Petrographic features and
thermal behaviour of the historically known ‘‘pietra ollare’’ from the Italian
Central Alps (Valchiavenna and Valmalenco). In: Maggetti, M., Messiga, B.
(Eds.), Geomaterials in Cultural Heritage, 257. Geological Society of London, pp.
229–239. Special Publication.
Bolla, M., 1991. Recipienti in pietra ollare. In: Caporusso, D. (Ed.), Scavi MM3.
Ricerche di archeologia urbana a Milano durante la costruzione della linea 3
della metropolitana 1982–1990, 3.2. I reperti, pp. 11–37. Milano.
Bonazza, A., Lazzarini, L., Vaccaro, C., 1999. Nuovo contributo archeometrico allo
studio della ‘‘pietra ollare’’ padana. In: 6 Giornata ‘Le scienze e l’archeometria’,
Este, pp. 171–180.
Bucher, K., Frey, M., 2002. Petrogenesis of Metamorphic Rocks. Springer-Verlag,
Berlin, 341 pp.
2501
Diderot, D., 1782. Encyclopédie, ou Dictionaire raisonné des sciences, des arts et des
métiers. Lausanne-Berne Ed.
Dietrich, V.J., 1980. The distribution of ophiolites in the Alps. Ofioliti, Special issue
on Thetyan ophiolites, 1, Western Area, pp. 7–51.
Donati, B., 1986a. La pietra ollare in Valmaggia dalla cava al laboratorio. In: 2000
anni di pietra ollare. Quaderni d’informazione, Bellinzona, pp. 17–42.
Donati, P., 1986b. Archeologia e pietra ollare nell’area Ticinese. In: 2000 anni di
pietra ollare. Quaderni d’informazione, Bellinzona, pp. 71–141.
Gaggi, S., Leoni, B., 1984. La pietra ollare. In: Quaderni della Provincia, Amminis-
trazione Provinciale di Sondrio.
Lusuardi Siena, S., Sannazaro, M., 1987. Pietra ollare di Luni, La pietra ollare in
Liguria. In: Atti della Giornata di Studio in ricordo di Lella Massari, Finale Ligure,
1985, pp. 165–198.
Malaguti, C., Zane, A., 1999. La pietra ollare nell’Italia nord-orientale. Archeologia
Medievale 26, 463–479.
Mannoni, T., Messiga, B., 1980. La produzione e la diffusione dei recipienti di pietra
ollare nell’Alto Medioevo. In: 6 Congresso Internazionale di studi sull’Alto
Medioevo, Milano, 1978. Proceedings, pp. 501–522.
Mannoni, T., 1986. Materia prima e scarti di produzione dei recipienti in pietra
ollare. Rivista di Studi Liguri 1–4 (LII), 155–164.
Mannoni, T., Pfeifer, H.R., Serneels, V., 1987. Giacimenti e cave di pietra ollare nelle
Alpi. In: La Pietra Ollare dalla Preistoria all’Età Moderna, Como, 1982.
Proceedings, pp. 7–45.
Mollo Mezzena, R., 1987. Primi elementi per lo studio della pietra ollare in valle
d’Aosta. In: La Pietra Ollare dalla Preistoria all’Età Moderna, Como, 1982.
Proceedings, pp. 59–114.
Mugnaini, D., 1995. Studio petroarcheometrico dei manufatti strumentali e orna-
mentali in serpentinite e clorito scisto di siti neolitici friulani e trentini. Thesys,
University of Bologna, 110 pp.
Pfeifer, H.R., Serneels, V., 1986. Inventaire des gisements de pierre ollaire
au Tessin et dans les regions voisines: aspects mineralogiques et miniers. In:
2000 anni di pietra ollare. Quaderni d’informazione, Bellinzona, pp. 147–235.
Robertson, E.G., 1982. Physical properties of building stone. In: Conservation of
Historic Stone Buildings and Monuments, Report of the Committee on
Conservation of Historic Stone Buildings and Monuments. National Academy
Press, Washington, D.C., pp. 62–86.
Santi, P., Antonelli, F., Renzulli, A., 2005. Provenance of Medieval pietra ollare
artefacts found in archaeological sites of Central-Eastern Italy: insights into the
Alpine soapstone trade. Archaeometry 47 (2), 253–264. Special issue: the
linking role of the Alps in past cultures: an archaeometric approach.