Journal of Archaeological Science 35 (2008) 1074e1089

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

Integrating archaeology, archaeometry and geology: local production

technology and imports at Paola (Cosenza, Southern Italy)
Elisabetta Gliozzo a,*, Paola Vivacqua b, Isabella Turbanti Memmi a
a
Department of Earth Sciences, University of Siena, via Laterina 8, 53100 Siena, Italy
b
Department of Archaeology and History of Arts, via Roma 52, 53100 Siena, Italy
Received 21 August 2006; received in revised form 26 July 2007; accepted 31 July 2007

Abstract

This paper documents a comprehensive study of coarse wares for table use, serving, washing and cooking, found at the archaeological site of
Paola (Cosenza, southern Italy). Major aims of the study were to group local products and imports and to determine technological features of local
products. Results were obtained through a combination of archaeological (stratigraphy and ceramic typology), chemical and mineralogicale
petrographical techniques (ICP-OES, ICP-MS, INAA, XRF, OM, SEM-EDS and EMPA). Data integration revealed that the coarse wares found
at Paola were produced locally or imported over short (Calabria region), medium (Campania, Pantelleria, Sicily) and long distances (Greece and
Africa). The local production focused on basins, stewpots, pitchers, bowls, pans, casseroles and lids, sometimes imitating African, or Pompei
(Campania) or San Foca (Puglia) typical shapes. The production technology of local ceramics used a sandyeclayey blend of materials collected
in the immediate vicinity of the archaeological site, fired at intermediate (about 800  C) to low (around 650e750  C) temperatures and used a
Fe-rich clayey compound as a red ceramic coating. From a technological perspective, the Paola example arise new questions about table wares
produced with clay bodies typical of cooking wares.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Coarse ware; Late Antique; Calabria; Local production and imports; Mineralogy and petrography

1. Archaeological and geological background on a hillside terrace overlooking the Paola site, which proba-
bly served as a coastal outlet where foodstuffs were stored
1.1. The archaeological area and small ships docked. The second period (3rde4th century
AD) witnessed the construction of a small building used as
The Roman and Late Antique settlement of Paola (Co- a thermal bath, subsequently replaced (third period, 4the5th
senza) lies within the present stadium, along the Tyrrhenian century AD) by a kiln. During the fourth period (5the7th cen-
coastal road (Fig. 1). From 1998 to 2000 three archaeological tury AD) the settlement grew to the size of a small village,
excavations were completed by the Soprintendenza Archeo- probably a Vicus. The excavation of quarters north of the small
logica of Calabria (scientific direction: Dott. G.F. Latorre; field bath brought to light another completely destroyed kiln, a sec-
director: Dott. A.B. Sangineto). The settlement covers about ond thermal bath and numerous homes and workshops. The
half an acre and spans four main periods (Lattanzi, 1999; San- settlement was inhabited to the end of the 6the7th century
gineto, 2000). The first period (early imperial age, 1ste2nd AD, when it was abandoned and used occasionally afterward
century AD) is represented by structures (maybe horrea) as a burial site.
likely linked to the villa at Contrada Cutura. The villa stands Although the specific function of the two kilns is uncertain,
tableware was likely produced in the smaller older one, while
* Corresponding author. Tel.: þ39 0577 233857; fax: þ39 0577 233938. the numerous Keay LII amphorae found at the site were fired
E-mail address: [email protected] (E. Gliozzo). in the larger more recent one.

0305-4403/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jas.2007.07.008
 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089 1075

Fig. 1. Geological map of the Paola territory (modified after Henderson and Page, 1963), where sediment sample points have been indicated.

1.2. The ceramic collection which became widespread in southern Italy starting in the 4th
century AD (Arthur, 1994).
Archaeological excavation of the settlement brought to Cultural and/or economic exchange did not occur only with
light numerous ceramic finds. A systematic study was con- southern Italy. Some ceramic fragments recall typical wares
ducted of the coarse wares (for table use, serving, washing from the Aegean (especially the stewpot type 9) and North
and cooking). One thousand six-hundred and twenty-nine frag- Africa (basin types 84 and 85), which were widespread in
ments from the 1st to 7th century AD were catalogued and di- the Mediterranean basin during the 5th and 6th centuries AD
vided into 104 types (Vivacqua, 2002). The varied and (see e.g. Fulford and Peacock, 1984; Bonacasa Carra, 1995).
complex forms show a high degree of standardization. Numer- The shape and fabric of the casseroles (types 25 and 26) are
ous types (especially some stewpots, pans, bowls, and basins) instead typical of ‘‘Pantellerian ware’’ (Peacock, 1982, 1984;
widely found in southern Italy suggest regional and interre- Santoro Bianchi et al., 2003). From a technological standpoint,
gional commercial trade. Among the cooking wares, the it is interesting to note that the ceramic bodies of some types
morphological homogeneity of some stewpots and pans is of tableware (e.g. pitcher type 64 and basin type 91) show typ-
noteworthy, with examples from Campania and Puglia (see ical fabrics of cookware.
e.g. Cotton, 1979; Carsana, 1994; Annese, 2000; Leone,
2000; Turchiano, 2000). The stewpot type 15, for example, 1.3. Paola geological settings
is comparable to the pottery from the 1ste2nd century AD
found at the S. Foca (Lecce) production site (D’Andria, The geology of Calabria, especially of its southern reaches, is
1980; Semeraro, 1992). Among the coarse wares for table similar not only to that of north-eastern Sicily, Sardinia, western
use, serving, and washing, instead, a few types of stewpots, Corsica, central-western Liguria but also to that of eastern por-
pitchers, cups and basins show traces of a red or white coating tions of Catalonia and Provençe, of central and north-eastern
(red: types 58, 69, 81.2, 86, 92; white: types 81.1, 84, 85, 88), Greece, of some Aegean Islands and of north-western Anatolia.
1076 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

These regions are all characterised by a prevalently acid alluvial sediments in outcrops along the San Francesco River.
metamorphic basement. Consequently, the mineralogicale CZ4 was sampled within alluvial deposits 400 m from the
petrographic composition of ceramics produced in these river, closer to the archaeological area. Unfortunately, recent
areas is so similar that it may be not easily diagnostic of construction work especially in the area around the archaeo-
provenance (D’Ambrosio et al., 1986, 1989). Nevertheless, logical site (which is surrounded by stadium bleachers) pro-
the detailed mineroepetrographic study of outcrops in the vi- hibited reliable sampling.
cinity of the archaeological site provides an important instru- Although the finds include numerous typologies, this study
ment for verifying the compatibility between local sediments focuses on coarse wares for table use, serving, washing and
and the examined artifacts. Based on published cartography cooking. A set of 55 samples (Table 1) was selected for this
(Henderson and Page, 1963) and on several research studies study. The criteria adopted for sample selection were the strati-
(Piccarreta, 1973; Bonardi et al., 1974; Lanzafame and Zuffa, graphic position of the find, ceramic typology and state of
1976) metabasites, marble and phyllarenites, phyllites and conservation. The stratigraphic sequence guided sampling, in
phyllarenites, muscoviteebiotite gneiss (and amphibolites), order to reconstruct the diachronic history of both local pro-
garnet micaschists, foliated granites, conglomerates, arenites duction and imports. The second selection criterion was based
and alluvial sediments crop out in the study area (Fig. 1). on the morphologicaletypological analysis of the vessels: we
As for the hydrography, the nearness of the coastal chain chose to sample the widest possible selection of identified
to the Tyrrhenian Sea is such that water channels are charac- types in order to obtain a heterogeneous assemblage. Although
terised by catchment basins of modest size and with a low this criterion no doubt complicates provenance studies, it aims
discharge strongly dependant on the annual precipitation. to provide a general overview of production in the complex,
The San Francesco River, which flows into the Tyrrhenian little-investigated region of Calabria. Lastly, we considered
about 1 km from the archaeological site of Paola, intersects the state of preservation, selecting the most intact samples rep-
all the previously mentioned outcrops. resentative of the various morphologies. No advanced degra-
dation was observed, except in the coatings.
2. Objectives
4. Sample preparation and analytical techniques
The current state of archaeological research has identified
a large number of ceramic productions in the Mediterranean Chemical analysis of pottery samples involved the applica-
basin, through the entire life of the site of Paola. Known pro- tion of different analytical techniques. Inductively Coupled
duction sites, however, are few and the majority of the produc- Plasma Optical Emission Spectrometry (ICP-OES) was used
tion systems must be reconstructed on the basis of the to determine major element and Ba, Sr, V, Y and Zr contents.
distribution of finished products. For this reason, the site of Inductively coupled plasma mass spectrometry (ICP-MS) re-
Paola is of the utmost importance because it has yielded kilns vealed Ag, Bi, Cd, Cu, Ni, Pb and Zn contents. Neutron acti-
for the production of numerous types of pottery and there has vation (INAA) was used to determine As, Au, Br, Co, Cr, Cs,
been a systematic study of the coarse wares. In this sense, ar- Hf, Ir, Rb, Sb, Sc, La, Ce, Nd, Sm, Eu, Tb, Yb, Lu, U and Th
cheometric study offers indicators for the typology of Paola’s contents. X-ray fluorescence (XRF) revealed Nb and Rb con-
ceramic production and provides chemical and mineroepetro- tents. Loss on ignition was determined by heating samples to
graphical references which allow the pottery to be recognized 1050  C for 2 h. See Gliozzo and Memmi (2004) for details
outside of its area of production. on sample preparation, instrumental characteristics and detec-
This study aims to determine the chemical and mineralog- tion limits. Principal Component Analysis and cluster analysis
ical characteristics of a collection of coarse wares from Paola were used to process chemical data. Principal Component
in order to distinguish local products and imported ones. The Analysis was based on the covariance matrix with no rotation
identification of artifacts locally produced is the most impor- of the axes and used major (wt%), minor and trace elements
tant objective. The further identification of local products (ppm) as variables; only P2O5, SO3 and LOI were excluded
(the territory of Paola and surroundings), regional (Calabria) because they were affected by post-depositional processes.
and Mediterranean contexts may provide information about The cluster analysis used the Average Linkage, based on the
patterns of imports or exports over short, medium and long quadratic Euclidean distance, and major, minor and trace ele-
distances. Another aim is to reconstruct production cycles at ments as variables (except for P2O5, SO3 and LOI). For clus-
the Paola kilns, the technological character of artifacts with ter analysis, the data proved to be particularly complex, as the
a red coating and the functional study of table wares produced numerous data analyses performed (e.g. using data in wt%
with clay bodies typical of cooking wares. and ppm and/or standardizing values) nucleated groups in
which the major elements presented standard deviations
3. Materials >3wt%. For this reason, minor and trace element contents
were all transformed from ppm to wt% in order to highlight
Four samples of sediments typically cropping out in the major element differences in chemical groupings and to use
area of Paola (Fig. 1) were analysed and compared with ce- the values of SiO2 and CaO as principle discriminating ele-
ramic finds in order to identify the source of locally fired ments, as indicated by PCA. Above all, the mineralogical in-
raw materials. CZ1, CZ2 and CZ3 are representative of the vestigation has demonstrated the absence of additives in the
 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089 1077

Table 1
Sample list
Sample Inv. US Shape Coating Type Type references Geographic diffusion of the vase type
PA 1 68 1 Stewpot e 1 Chiaromonte Trerè Southern Italy
PA 2 71 1 Stewpot e 1 (1984, no.1, pl. 101, no. 8e10, pl. 100) Southern Italy
PA 3 62 195 Stewpot e 2 Bonacasa Carra (1995, no. 85/323, Fig. 68) Central and southern Italy
PA 4 60 160 Stewpot e 3 Carsana (1994, no. 31.2e31.3, Fig. 111) Southern Italy
PA 5 61 2 Stewpot e 5 Fulford and Peacock (1984, no.29.3, Fig. 70) Central, southern Italy þ northern Africa
PA 6 66 169 Stewpot e 6.1 Annese et al. (2000, no. 1.1, pl. 7) Mediterranean (aegean origin?)
PA 7 70 1 Stewpot e 7.1 Carsana (1994, no. 33.2, Fig. 113) Southern Italy
PA 8 65 1 Stewpot e 9 Bonacasa Carra (1995, no. 69, pl. 22) Mediterranean (aegean origin?)
PA 9 64 150 Stewpot e 11 Small and Buck (1994, no. 73, Fig. 56) Southern Italy
PA 10 59 169 Stewpot e 12.1 Fulford and Peacock (1984, no. 19.2, Fig. 69) Central Italy þ northern Africa
PA 11 67 194 Stewpot e 13 Di Giuseppe (1998, no. 9, Fig. 9) Southern Italy
PA 12 72 130 Stewpot e 14 Small and Buck (1994, no. 17, Fig. 31) Central and southern Italy
PA 13 63 216 Stewpot e 15 D’Andria (1980, pp. 79e88) Mediterranean (southern Salento origin?)
PA 14 69 115 Stewpot e 17 Fracchia (1994, no. 43, Fig. 79) Southern Italy
PA 15 50 169 Pan e 18 Annese (2000, no. 9.3, pl. 17) Central and southern Italy
PA 16 48 170 Pan e 19 Carsana (1994, no.13, Fig. 109) Southern Italy
PA 17 46 115 Pan e 20 Fracchia (1994, no. 18, Fig. 77) Central and southern Italy
PA 18 49 194 Pan e 22 Hayes (1972, no. 12, Fig. 5) Mediterranean (copy of African)
PA 19 51 115 Casserole e 25 Bonacasa Carra (1995, no. 85e88, Fig. 71) Pantellerian ware
PA 20 52 130 Casserole e 26 Carsana (1994, no. 112.5, Fig. 121) Pantellerian ware
PA 21 26 170 Casserole e 30.1 Hayes (1972, no. 181, Fig. 35) Southern Italy (copy of African)
PA 22 27 195 Casserole e 32 Annese (2000, no. 1.8, pl. 16) Mediterranean
PA 23 30 115 Casserole e 33 Ostia III (no.700, pl. 74) Mediterranean
PA 24 29 115 Casserole e 34 Whitehouse et al. (1985, no. 72, Fig. 6) Mediterranean
PA 25 56 130 Lid e 36 Annese (2000, no. 28.3, pl. 22) Southern Italy
PA 26 58 115 Lid e 37.1 Annese (2000, no. 28.1, pl. 21) Southern Italy
PA 27 57 194 Lid e 39 Carsana (1994, no. 72.2, Fig. 116) Mediterranean (copy of African)
PA 28 55 130 Bowl e 48 Hayes (1972, no. 99, Fig. 28, no. 61B, Fig. 17) Mediterranean
PA 29 53 170 Bowl e 49 Hayes (1972, no. 103A, Fig. 29) Mediterranean (copy of African)
PA 30 19 42 Bowl e 51 Hayes (1972, no. 97, Fig. 27) Mediterranean (copy of African)
PA 31 05 207 Cup e 52 Annese (2000, no. 2.1, pl. 10) Central and southern Italy
(copy of African)
PA 32 04 2 Cup e 53.2 Fracchia (1994, no. 32, Fig. 79) Southern Italy þ northern Africa
PA 33 43 1 Pitcher Red 58 Arthur (1994, no. 99, Fig. 95) Mediterranean
PA 34 40 115 Pitcher e 60 Fracchia (1994, no. 5, Fig. 77) Mediterranean
PA 35 45 194 Pitcher e 62 Fulford and Peacock (1984, no. 33.1, Fig. 81) Central Italy þ northern Africa
PA 36 37 150 Pitcher e 64 Annese et al. (2000, no. 12.1, pl. 3) Southern Italy þ northern Africa
PA 37 36 130 Pitcher e 65.1 Small and Buck (1994, no. 66, Fig. 54) Mediterranean
PA 38 38 194 Pitcher e 65.2 Leone (2000, no. 9.1, pl. 9) Southern Italy
PA 39 39 169 Bottle e 67 Ostia III (no. 696, pl. 73) Central and southern Italy
PA 40 32 2 Stewpot Red 69 Small and Buck (1994, no. 153, Fig. 114) Southern Italy
PA 41 34 130 Stewpot e 70 Di Giuseppe, (1998, no. 2, Fig. 7) Mediterranean
PA 42 33 228 Stewpot e 71 Fracchia (1994, no. 14, Fig. 77) Southern Italy
PA 43 22 215 Basin e 75 Ostia III (no. 270c, pl. 37) Central and southern Italy
PA 44 54 1 Basin e 79 Fulford and Peacock (1984, no. 20.1, Mediterranean
Fig. 58, no. 32.1, Fig. 59)
PA 45 31 194 Basin White 81.1 Bonacasa Carra (1995, no. 85.158, Fig. 42) Mediterranean
PA 46 73 115 Basin Red 81.2 e e
PA 47 15 224 Basin White 84 Arthur (1994, no. 68, Fig. 90) Southern Italy (african origin?)
PA 48 16 224 Basin White 85 Cotton (1979, no. 15, Fig. 56) Central and southern Italy
(african origin?)
PA 49 09 194 Basin Red 86 Bonacasa Carra (1995, no. 86.35, Fig. 45) Mediterranean
PA 50 24 224 Basin White 88 Leone (2000, no. 2.1, pl. 7) Southern Italy
PA 51 23 42 Basin e 90 Fulford and Peacock (1984, no. 17.2, Fig. 63) Central, southern Italy þ northern Africa
PA 52 08 42 Basin e 91 Fulford and Peacock (1984, no. 1.2, Fig. 76) Northern Africa
PA 53 12 1 Basin Red 92 Arthur (1994, no. 16.4, Fig. 81) Southern Italy þ northern Africa
PA 54 21 1 Basin Red 93 Hayes (1972, no. 3, Figs. 67e69) Southern Italy
PA 55 42 1 Basin e 94 Annese (2000, no. 6.2, pl. 11) Southern Italy þ northern Africa
US, findsite stratigrafic unit.
In ‘‘type references’’ column, only one reference e providing the best comparison for the vessel morphology e has been listed.
1078 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

clay, and thus, the contents of SiO2 and CaO must be repre- can be established (Table 3). CG 1e12 comprise 50 samples
sentative of the natural sediment. With respect to the minor with less than 10 wt% CaO, while CG 13e16 comprise 9
and trace elements, they show rather large standard deviations with 12e15 wt% CaO. There is apparently no correlation be-
in all of the studies but the best results were obtained using tween CaO and MgO, with the latter showing comparatively
two different procedures: not standardizing the values and us- higher values in CG 2, 10e12, 15 and 16. The SiO2 content
ing all data in wt% (proposed here), or standardizing values is high (or intermediate) in CG 1e3 and 5e8, whereas it is in-
between the 0e1 range and using data in wt% and ppm (sac- termediate to low in the other groups. CG 9, for example,
rificing, however, the compositional homogeneity of the major shows the lowest SiO2 contents which are compensated by
elements). The dendogram proposed below presents groups particularly high Al2O3 and K2O contents. Na2O reaches
with minor values at a standard deviation, both in terms of 4 wt% in CG 4, which is also characterised by high TiO2,
major or minor and trace elements; above all, contrary to Al2O3, Fe2O3 and K2O contents. Fe2O3 and MnO contents
all other analyses, this is compatible with groupings nucleated are highest in CG 11 samples, whereas TiO2 shows no signif-
by petrographic analysis and with matrix chemical composi- icant variations. CG 1e6, 9e11 (maybe also 12) maybe distin-
tions obtained by SEM-EDS. guished from the other groups for the high contents of most
Textural and mineralogical investigations were conducted minor and trace elements. In particular, note the high Ni and
by optical microscopy (OM). Results from the statistical anal- Cr contents in sample PA 8, and the high Zr and Ba contents
ysis of bulk chemical data and from OM investigations guided in CG 4 specimens.
the selection of key samples to be studied under the scanning
electron microprobe (SEM-EDS). Instrumentation consisted of 5.2. OM, SEM and EMPA investigations
a Philips XL30 SEM equipped with a Philips EDAX DX4 en-
ergy dispersive spectrometer (EDS). A few ceramic samples of Sediments: the grain size distribution of sediments is that of
major interest were further investigated by electron micro- a clayey sand. The mineralogical composition of sediments is
probe analysis (EMPA) using a CAMECA SX50 (IGG at the characterised by the presence of abundant quartz (sometimes
University of Padova). Operating conditions were: 15 kV, showing undulatory extinction), fresh or saussuritized plagio-
beam current at 15 mA. PAP software was used for correction. clase (albite to bytownite in composition), K-feldspar (some-
Sediment compositions were investigated using OM, SEM- times perthitic or sericitic), muscovite, frequently chloritized
EDS and XRF (Philips MagixPro). biotite, compositionally variable Mg/Al to Mg/Fe chlorites,
almandineepyrope and Feehornblende. Microcline, augite,
5. Results grossular and epidote ss. are occasionally present. Calcite, do-
lomite, actinolite, cummingtonite, enstatite and pumpellyite
5.1. Bulk chemical analyses are rare, while lawsonite, cordierite and sillimanite are
extremely rare. Accessory minerals include ilmenite, titanite,
Sediments: chemical compositions are reported in Table 2. apatite, zircon and iron sulphide.
CaO contents are variable but generally less than 9 wt%. Al- Ceramics: the entire assemblage was divided into 9 miner-
kali contents are relatively high, whereas the average SiO2 alogicalepetrographic (MPG) groups, based on matrix and
content is 56 wt%. Minor and trace element contents are vari- skeleton composition. For the sake of clarity the main charac-
able and not discriminating. teristic of the groups has been reported in Table 4. Group by
Ceramics: bulk chemical data are reported in Table 2. Sam- group, the correspondence between mineralogicalepetro-
ples show variable CaO contents: 40 specimens contain less graphic and chemical groupings are here indicated, together
than 5 wt%, 10 contain 5e10 wt%, and 9 contain 10e with additional observations on present phases, crystal habits
16 wt%. The ample variations in SiO2, Al2O3, Fe2O3, Na2O and chemical composition.
and K2O contents, and in all minor and trace elements, are The MPG 1 includes samples clustered in the chemical
such that statistical processing was required. As far as PCA groups 1 and 2 (CG 1e2). In all samples, quartz sometimes
is concerned, the first three principal components account shows an undulatory extinction, plagioclase (albite to andesine
for 90% of the total variation, with the first, second and third in composition) may be fresh or saussuritized, K-feldspar is
principal components accounting for 51.97%, 27.09% and sometimes sericitic, biotite is Al-rich and K-poor, garnet is al-
10.15% of variation, respectively. The two-dimensional com- mandineepyrope, and the FeeMgehornblende is particularly
ponent plot in Fig. 2A shows the variables, labelled as ele- titaniferous in sample PA 7. Further minerals are rare grossular
ments, for the first two principal components. SiO2 affects in PA 1 and 7, microcline in PA 7, crystals with myrmekite and
the first principal component and is negatively correlated cummingtonite in PA 1, and actinolite and extremely rare au-
with CaO, MgO and Sr. TiO2, Al2O3, Fe2O3, K2O and gite in PA 8 and 40.
Na2O, along with most minor and trace elements, affect the With the exception of sample PA 29, the MPG 2 includes
second principal component (positive) and do not represent samples clustered in the chemical group 3 (CG 3). In these lat-
significant discriminating variables for the examined assem- ters, plagioclase is albite to labradorite in composition, K-feld-
blage. Cluster analysis (Fig. 2B) reveals significant heteroge- spar is sometimes microperthitic and chlorite is Al-poor.
neity within the selected sample-set; nevertheless, 16 Sample PA 24 contains phyllite fragments with a coarser grain
chemical groups (CG) with acceptable standard deviations size (maybe phyllitic schists).
Table 2
Chemical analysis performed both on ceramic samples and sediments as received
Sample SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 SO3 LOI Sc V Cr Co Ni Cu Zn As Rb Sr Y Zr Sb Cs Ba La Hf Pb Ce Nd Sm Eu Tb Yb Lu Th U
CZ 1 53.79 1.07 16.97 7.50 0.13 3.18 4.05 1.71 2.48 2.48 e 8.66 e 146 104 25 48 e 118 e 101 192 27 214 e e 531 37 e 32 74 e e e e e e 9 4
CZ 2 58.84 0.75 19.12 6.42 0.11 4.31 8.33 1.87 1.37 1.37 e 4.86 e 128 135 28 48 e 61 e 43 255 24 135 e e 356 16 e 14 28 e e e e e e 12 2
CZ 3 58.72 0.87 18.64 6.62 0.19 2.02 1.91 1.73 3.17 3.17 e 5.23 e 116 77 26 34 e 88 e 128 159 24 206 e e 662 36 e 29 85 e e e e e e 6 4
CZ 4 54.88 1.19 18.03 7.37 0.15 5.05 4.15 2.27 2.35 2.35 e 4.23 e 147 112 31 52 e 68 e 86 215 25 188 e e 465 27 e 21 36 e e e e e e 8 3
PA 01 62.38 0.88 18.64 6.93 0.274 1.80 1.33 1.66 3.07 0.17 0.005 2.91 15 97 89 35 62 47 123 13 132 163 32 275 1 6 501 62 7 46 117 51 9 2 1 4 0.6 17 2
PA 02 61.35 0.82 18.31 6.89 0.043 0.84 1.82 1.30 1.81 0.31 0.004 6.54 17 121 70 10 23 39 73 13 75 205 21 251 1 5 687 38 6 40 69 31 5 1 1 2 0.4 15 4
PA 03 57.45 0.89 17.81 5.05 0.036 0.89 2.50 1.85 2.04 0.25 0.002 11.19 14 96 51 9 15 33 83 8 92 278 26 253 1 3 706 69 7 36 128 50 9 2 1 3 0.4 14 3
PA 04 58.87 1.01 20.15 8.23 0.079 1.89 1.37 1.05 2.87 0.28 0.005 4.23 20 106 138 34 73 49 123 13 127 176 42 236 1 7 476 81 6 30 168 56 12 2 2 4 0.6 14 4
PA 05 58.25 0.68 17.51 5.23 0.013 0.55 1.38 1.15 1.92 0.22 0.011 13.12 11 80 72 5 25 18 63 16 68 159 15 321 1 8 570 30 7 36 61 26 4 1 0 2 0.3 25 4
PA 06 61.82 1.04 21.00 8.10 0.113 1.60 1.37 1.20 2.91 0.11 0.002 0.75 21 131 115 17 51 38 104 18 185 138 27 262 1 8 473 51 6 29 109 39 7 2 1 3 0.5 16 4

E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

PA 07 66.54 0.75 18.57 6.09 0.082 1.47 0.78 1.39 2.70 0.15 0.012 1.83 14 105 72 12 31 29 89 15 120 111 15 247 1 10 462 38 5 31 59 30 5 1 0 2 0.3 19 4
PA 08 64.70 0.94 17.57 8.27 0.096 2.19 0.91 1.24 1.49 0.10 0.002 2.44 18 144 277 31 214 35 105 20 84 64 32 240 2 5 306 36 6 36 70 33 6 1 1 4 0.5 10 4
PA 09 62.04 0.87 19.14 7.22 0.086 3.58 0.72 0.87 3.48 0.09 0.007 1.91 19 135 106 25 61 49 81 26 142 71 40 222 1 11 402 49 5 19 86 44 9 2 1 4 0.6 16 3
PA 10 59.53 0.87 18.13 7.25 0.120 1.36 3.94 1.71 2.24 0.48 0.008 4.35 15 106 67 21 30 72 108 8 131 475 26 201 1 5 738 56 5 31 106 48 8 2 1 3 0.4 14 4
PA 11 59.34 0.664 18.67 6.37 0.055 0.96 1.74 1.85 2.43 0.36 0.006 7.87 14 97 60 15 26 62 78 12 101 225 23 194 1 4 647 47 5 25 95 39 7 2 0 2 0.3 16 3
PA 12 56.95 0.92 16.78 5.40 0.059 1.05 2.32 1.86 2.01 0.16 0.013 11.88 15 95 44 11 18 32 66 13 90 254 23 313 1 5 554 67 9 26 125 54 9 2 0 2 0.4 24 3
PA 13 58.96 0.70 16.07 5.87 0.055 1.05 1.24 1.11 2.01 0.42 0.008 12.52 15 104 148 17 77 26 72 17 62 99 28 183 1 3 478 39 5 28 84 29 6 1 1 3 0.4 13 3
PA 14 63.75 0.63 16.59 4.94 0.265 0.97 1.27 1.61 2.72 0.28 0.006 7.04 12 94 78 36 35 31 61 11 86 179 27 188 1 5 554 51 5 24 93 43 8 2 1 3 0.4 14 3
PA 15 58.31 1.07 21.29 9.26 0.171 1.16 2.27 1.70 1.72 0.24 0.008 2.63 21 173 66 27 29 90 103 13 70 372 32 235 1 5 673 67 5 25 124 53 9 2 1 3 0.5 15 4
PA 16 62.74 0.80 16.95 6.58 0.177 0.85 1.40 1.22 1.86 0.32 0.017 6.66 12 135 116 26 71 39 83 31 83 254 38 406 2 13 944 117 9 102 236 85 13 3 1 3 0.5 50 4
PA 17 64.63 0.66 17.44 5.40 0.117 1.04 1.55 1.35 2.76 0.16 0.009 4.78 12 76 81 10 31 25 77 15 106 127 18 213 1 7 541 35 5 30 59 27 5 1 0 2 0.3 15 3
PA 18 64.80 0.72 15.64 5.28 0.081 0.68 1.43 1.24 2.38 0.33 0.008 6.18 11 67 77 8 25 29 79 11 96 143 19 267 1 9 530 40 6 34 82 38 5 1 0 2 0.3 19 3
PA 19 52.15 0.98 19.59 7.08 0.130 0.52 2.10 3.90 2.37 0.35 0.007 10.86 8 12 13 5 0 19 138 6 49 292 64 668 0 0 1109 96 14 16 180 75 14 4 2 6 1.0 11 2
PA 20 55.04 0.82 19.25 6.62 0.250 0.57 1.62 4.35 2.73 0.28 0.005 8.63 7 24 10 4 7 16 143 8 43 243 56 683 0 2 1062 90 13 21 178 65 13 5 2 6 0.8 15 3
PA 21 48.37 0.93 16.38 6.56 0.419 2.51 12.70 1.20 1.33 0.73 0.019 8.85 17 105 135 23 57 38 91 11 56 547 28 134 1 2 388 39 4 20 71 25 6 1 1 3 0.4 10 3
PA 22 49.98 1.13 18.28 9.08 0.307 2.92 2.08 1.32 1.86 0.36 0.005 12.57 21 121 177 42 87 54 160 13 83 170 33 238 1 4 954 58 6 43 111 51 10 2 1 4 0.6 15 2
PA 23 61.34 0.61 15.69 4.73 0.094 1.20 1.82 1.72 2.95 0.33 0.008 9.48 12 78 68 8 23 42 82 8 100 165 19 183 1 4 643 36 4 26 66 30 5 1 0 2 0.3 12 2
PA 24 60.47 0.67 16.79 5.37 0.234 1.29 1.28 1.27 2.40 0.25 0.012 9.94 12 116 69 31 37 38 136 16 90 160 14 215 2 7 533 33 5 61 51 21 4 1 0 1 0.2 19 3
PA 25 57.86 1.02 17.21 6.02 0.061 0.94 2.26 1.58 2.11 0.32 0.002 10.65 15 103 50 10 18 47 81 12 113 282 21 376 1 4 601 56 9 32 101 54 7 1 1 3 0.4 17 3
PA 26 57.71 0.75 16.49 5.74 0.049 0.81 1.49 0.75 1.53 0.34 0.007 13.31 13 106 85 8 28 46 72 17 68 160 21 233 1 9 449 34 6 32 61 27 5 1 0 2 0.3 17 3
PA 27 45.69 0.79 21.57 6.61 0.173 0.77 2.56 1.39 3.06 1.00 0.006 16.37 10 88 56 14 31 34 127 29 148 411 43 369 2 18 842 107 8 91 179 73 12 2 1 4 0.7 27 5
PA 28 69.11 0.53 11.17 5.06 0.057 1.19 6.29 0.55 1.85 0.42 0.024 4.14 10 71 92 11 31 20 34 21 81 136 34 268 2 6 273 36 6 14 70 34 7 2 1 3 0.4 10 4
PA 29 63.18 0.69 16.59 5.72 0.125 1.03 1.56 1.62 2.81 0.28 0.005 6.44 11 105 84 15 39 32 92 12 121 202 29 256 1 7 552 60 6 34 122 38 8 2 1 3 0.5 16 5
PA 30 62.56 0.70 12.07 4.69 0.071 1.36 9.73 0.51 1.26 0.65 0.014 6.42 11 108 97 11 25 14 37 7 39 274 21 186 1 2 343 35 5 16 68 28 5 1 1 2 0.4 8 2
PA 31 55.54 1.04 18.51 7.00 0.038 1.04 1.57 1.02 1.87 0.70 0.002 11.59 17 88 90 8 26 35 86 14 92 155 22 284 1 6 614 51 7 55 95 37 7 1 1 3 0.5 23 4
PA 32 57.33 1.39 23.04 8.47 0.080 1.59 1.07 1.78 3.11 0.16 0.003 2.14 21 115 100 15 38 39 116 23 162 101 41 371 2 9 631 87 10 40 138 61 11 2 2 4 0.6 29 4
PA 33 69.01 0.73 12.24 5.41 0.047 1.27 3.68 0.42 1.36 0.43 0.010 5.38 11 55 99 12 28 21 86 7 50 169 22 270 1 3 244 37 6 21 68 24 5 1 1 2 0.3 9 2
PA 35 54.31 0.86 15.39 6.16 0.114 1.83 5.03 0.60 1.86 0.97 0.035 12.52 15 82 110 15 43 28 95 15 53 197 30 207 1 5 391 40 4 26 80 38 7 2 0 3 0.5 12 3
PA 34 55.25 0.90 17.01 7.04 0.071 3.33 7.51 1.30 2.67 0.82 0.012 4.03 19 95 160 18 86 47 148 5 96 328 31 176 1 6 455 46 4 26 83 38 7 1 1 3 0.5 13 2
PA 36 65.10 0.80 19.88 6.49 0.074 1.39 0.90 1.19 2.87 0.12 0.005 1.19 15 122 82 10 31 25 75 18 124 95 15 312 2 13 444 38 7 33 61 28 4 1 0 2 0.3 22 4
PA 37 50.12 0.92 16.96 7.14 0.160 2.76 6.20 1.05 2.73 0.54 0.018 11.35 18 90 141 19 81 57 144 13 124 240 27 160 1 6 485 43 4 24 87 38 7 1 1 3 0.5 13 2
PA 38 54.64 0.78 16.01 6.47 0.063 3.36 13.45 0.88 2.30 0.28 0.007 1.45 18 123 153 16 75 41 103 4 70 357 25 140 1 8 371 36 4 11 69 32 5 1 0 3 0.4 12 2
PA 39 49.67 0.62 11.77 5.92 0.158 2.07 15.92 0.80 1.77 0.22 0.049 10.45 14 89 243 22 122 25 68 21 67 469 28 192 2 6 361 37 5 22 69 35 6 1 1 3 0.4 11 3
PA 40 63.11 0.90 20.27 7.45 0.034 2.43 0.88 1.11 2.25 0.14 0.007 1.40 19 151 150 15 62 30 120 21 115 138 32 203 2 15 499 51 5 28 86 39 7 2 0 3 0.5 15 4
PA 41 49.94 0.80 15.34 6.04 0.237 2.23 12.16 1.08 2.21 0.73 0.060 9.21 17 74 104 18 50 50 84 11 79 601 30 172 2 3 638 49 4 27 91 38 7 1 1 3 0.5 12 4
PA 42 51.88 0.70 14.03 5.73 0.107 2.64 10.52 0.91 1.80 0.61 0.031 11.03 15 93 125 14 48 54 99 11 73 339 23 140 1 3 409 33 3 26 64 28 6 1 1 2 0.4 9 2
PA 43 60.52 0.67 17.81 5.14 0.012 0.45 1.04 1.19 2.02 0.14 0.006 10.93 10 76 60 5 24 16 60 16 87 126 17 325 1 7 545 33 8 28 48 29 4 1 0 2 0.2 23 4

1079
(continued on next page)
1080 La Hf Pb Ce Nd Sm Eu Tb Yb Lu Th U E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

The MPG 3 includes samples clustered in the chemical

4
2
3
3
2
2
3
2
2
2
4
5
0.4 16 3
7 1
0.2 7 2
1 50 5
36

11
11
13
12

10
19
19
7

9
7
8
group 4 (CG 4) and shows a mineral assemblage constituted
0.6
0.3
0.4
0.5
0.4
0.4
0.4
0.3
0.4
0.4
0.3
0.5
by anorthoclase, aenigmatite, aegirine, fayalite and glass. Ac-

Oxides expressed in wt%, minor and trace elements expressed in ppm. Mean, standard deviation (S.D.), minimum and maximum values for the whole ceramic sample-set have been further displayed.
0
cessory minerals include rare ulvöspinel and extremely rare
4
2
3
4
3
2
3
2
3
2
2
3
3
1
1
6
quartz.
2
1
1
1
1
0
1
1
0
1
0
1
1
1
0
2
The MPG 4 includes samples clustered in the chemical groups
51 179 67 11 3
27 54 24 5 1
8 71 32 6 1
28 82 31 6 2
18 69 36 7 1
18 54 30 6 1
18 70 26 5 1
31 55 22 4 1
36 68 27 5 1
19 58 31 5 1
25 64 27 5 1
33 86 31 5 1
1
1
1
5
5 and 6 (CG 5 and 6), with the exception of sample PA 15. The

85 16
39 7
14 3
21 4
mineralogicalepetrographic assemblage shows a combination
of the characteristics observed in MPG 1 and 2. Plagioclase is al-

19 1109 117 14 119 236

6 518 49 6 31 93
4 193 21 2 16 40
8 48
bite to andesine in composition in samples PA 5 and 52, and albite
to labradorite in samples PA 3 and 32. Biotite is ubiquitous and
frequently chloritized; almandineepyrope is particularly abun-
36 10
90 8
28 5
37 3
42 7
41 8
36 3
37 8
26 4
27 4

40 8
40 8

0 244 26 2
dant in samples PA 3 and 32. PA 32 shows an alumosilicate inclu-
sion (too small to be identified with certainty, maybe sillimanite)
548
277
441
338
369
349
305
314
293
329
558
334

within an almandine crystal, and cordierite associated with al-

Sb Cs Ba

mandine and chlorite in a lithic fragment. All samples further con-

19

10
1
8
4
4
2
4
3
4
4

tain abundant Fe and Ti oxides and zircon.

The MPG 5 includes samples PA 29 and 15, respectively,
2
1
1
1
1
1
1
2
1
1
1
1
1
0
0
2
398
245
122
282
295
132
331
157
158
396
271
351
36 93 14 91 237 27 260
15 27 6 36 124 9 109

90 292 31 200 883 68 683

14 34 4 28 64 14 90

clustered in the chemical groups 3 and 5 (CG 3 and CG 5). In

Y Zr

these two samples, kaersutite, biotite, spessartine-almandine,

37
19
27
27
28
26
24
18
22
25
18
28

olivine and glass occur. Olivine is compositionally intermediate

11 22 24 100 30 200 376
12 28 20 70 12 31 288
19 55 34 85 9 85 534
14 40 27 101 20 68 255
15 42 26 98 17 71 206
14 54 35 96 13 62 303
9 25 16 69 16 62 287
12 31 59 92 14 28 348
31 311 38 73 18 41 133
10 30 18 105 15 71 208
9 29 24 77 17 125 112
7 22 24 71 12 108 117
Cu Zn As Rb Sr

to forsteriteefayalite (Fo38).
Glass is the main constituent of samples included in the
MPG 6, clustered in the chemical group 9 (CG 9). In these
samples, sanidine, plagioclase (bytownite in composition), sal-
ite rare grossular and salite crystals and extremely rare crys-
tals of forsteritic olivine (Fo78) are also present, the extremely
21 173 411 57 311
14 96 105 16 49
4 29 63 9 48
0
Co Ni

rare quartz crystals are always less than 15 mm in diameter.

7 12 10 4

The MPG 7 and 8 include samples clustered in the chemi-

143
123

119

411
40
82

99

97
75

77
75
86

cal groups 7 (CG 7) and 12 (CG 12), respectively. It is worth

Cr

noticing that quartz contained in MPG 8 samples is frequently

17 136

13 109
9 87
11 48
17 97
14 90
13 84
14 79
11 67
11 61

10 75

11 65
Sc V

rounded.
Lastly, the MPG 9 includes samples clustered in the chem-
9.51
8.56

4.46

3.73
1.11
4.78
14.15
15.36

12.09
13.96

20.17
10.93

7.9
4.6
0.8
LOI

ical groups 13, 14 and 15 (CG 13e15). In these samples, pla-

25

gioclase is albite to andesine in composition, K-feldspar is less

0.005
0.054
0.035
0.020
0.014
0.044
0.018
0.047
0.005
0.022
0.006
0.004
0.01
0.01
Sample SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 SO3

abundant than plagioclase and biotite is sometimes chloritized.

0.1
0

Two minute lithic fragments in PA 46 show the following as-

0.34
0.55
0.88
0.38
0.39
0.85
0.26
0.38
0.24
0.29
0.10
0.14
0.4
0.2
0.1
1.0

sociation: (1) quartz, K-feldspar, chloritized biotite and pum-

pellyite; (2) albite/oligoclase and pumpellyite. Dolomite and
2.79
0.86
2.02
1.80
1.60
1.93
1.73
0.93
0.83
2.05
2.99
1.72
2.2
0.6
0.8
3.5

augite are occasionally present.

1.24
0.30
0.87
0.57
0.53
0.76
0.59
0.38
1.65
0.36
1.30
1.38

Sample PA 38 (CG 16) stands out for its advanced degree

1.2
0.7
0.3
4.4

of vitrification which greatly hampers the detection and iden-

2.35

9.13
6.42

5.94

1.66
4.93
1.13
0.74
13.79
12.62

10.47

14.54

4.4
4.4
0.7
27.1

tification of crystals. Rare quartz, potassic alkali-feldspar crys-

tals and Fe and Ti oxides were observed.
0.80
0.89
2.30
1.54
1.24
2.48
1.10
0.90
2.07
1.58
1.25
0.74
1.5
0.8
0.5
3.6

Matrix compositions were obtained through 10 SEM-EDS

raster analyses in each of the 23 selected samples. Data
0.131
0.040
0.130
0.076
0.075
0.116
0.029
0.056
0.067
0.044
0.053
0.022
0.11
0.08

0.4

(Table 5) indicate that ferric matrices predominate with

0

respect to carbonate ones; CG 5 specimens stand out for their

6.44
4.39
6.02
6.28
6.31
5.79
4.97
4.73
7.61
5.00
5.90
5.96
6.3
1.1
4.0
9.3

high FeO contents, sample PA 27 (CG 9) for its high MnO

content and sample PA 39 (CG 13) for its particularly high
21.76
10.36
15.52
14.55
14.81
15.25
12.35
10.49
15.39
11.97
18.41
17.84
16.9

10.4
23.0
2.9

CaO content.
EMP analyses (Table 6) were completed on the aenigmatite
0.76
0.59
0.81
0.90
0.89
0.69
0.75
0.57
0.76
0.77
0.76
0.80
Table 2 (continued)

0.8
0.2
0.5
1.4

crystals (Fig. 3A) and glass fragments (Fig. 3B) of the MPG 3
49.15
52.63
49.34
56.24
55.53
47.95
67.82
46.87
58.82
69.39
66.93
65.82

samples, on the kaersutite (Fig. 3C), biotite, olivine, spes-

58.2

27.7
69.4
6.3

sartineealmandine crystals and glass fragments (Fig. 3D) of

44
45
46
47
48
49
50
51
52
53
54
55

the MPG 5 samples, on the sanidine, salite (Fig. 3E), bytownite

Mean

Max.
Min.
S.D.
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA

crystals and glass fragments (Fig. 3F) of the MPG 6 samples.

Fig. 2. Statistical treatment of chemical data. (A) PCA two-dimensional component plot. (B) Dendrogram from cluster analysis. [T, type; CG, chemical group; c.,
coating].
 1082
Table 3
Mean and standard deviation values for major, minor and trace elements content of the 16 chemical groups indicated by cluster analysis [CG, chemical group]
CG SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O Sc V Cr Co Ni Cu Zn As Rb Sr Y Zr Ba La Hf Pb Ce Nd Sm Eu Tb Yb Lu Th U
1 (n ¼ 5) 65.8 0.8 18.5 6.5 0.1 1.4 0.9 1.3 2.4 14 109 118 14 65 27 83 16 112 100 22 284 421 38 7 32 68 30 5 1 0 3 0 18 4
d.s. 0.9 0.1 0.9 1.0 0 0.5 0.2 0.1 0.7 3 29 89 10 83 5 14 3 17 22 8 47 102 2 1 4 11 2 1 0 1 1 0 5 0
2 (n ¼ 4) 62.3 0.9 19.8 7.4 0.1 2.4 1.1 1.2 2.9 19 129 115 23 59 41 107 20 144 128 33 241 469 53 6 31 100 43 8 2 1 4 1 16 3

E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

d.s. 0.6 0.1 1.1 0.5 0.1 0.9 0.3 0.3 0.5 3 23 26 9 5 9 19 5 30 39 5 34 46 6 1 11 16 6 1 0 1 1 0 1 1
3 (n ¼ 9) 62.5 0.7 16.9 5.6 0.1 0.9 1.5 1.4 2.4 12 96 78 17 34 32 83 15 94 173 22 256 614 49 6 42 92 38 6 1 0 2 0 20 3
d.s. 1.7 0.1 1.1 0.8 0.1 0.4 0.4 0.2 0.4 3 21 13 10 11 10 29 3 11 32 3 53 84 4 1 14 13 5 1 0 0 1 0 4 1
4 (n ¼ 2) 53.6 0.9 19.4 6.9 0.2 0.5 1.9 4.1 2.6 8 18 12 5 4 18 141 7 46 268 60 676 1086 93 14 19 179 70 14 5 2 6 1 13 3
d.s. 2.0 0.1 0.2 0.3 0.1 0 0.3 0.3 0.3 1 8 2 1 5 2 4 1 4 35 6 11 33 4 1 4 1 7 1 1 0 0 0 3 1
5 (n ¼ 3) 58.2 1.2 21.5 8.7 0.1 1.5 1.6 1.5 2.6 21 131 101 25 47 59 114 16 120 216 38 281 593 78 7 32 143 57 11 2 2 4 1 19 4
d.s. 0.8 0.2 1.5 0.5 0.1 0.4 0.6 0.4 0.7 1 36 36 10 23 27 10 6 46 140 6 78 104 10 3 8 22 4 2 0 1 1 0 8 0
6 (n ¼ 10) 58.0 0.8 17.3 6.2 0.1 1.1 2.0 1.5 1.9 15 101 108 14 57 41 78 14 86 222 23 252 565 48 6 34 92 39 7 1 1 3 0 17 3
d.s. 1.2 0.1 1.1 0.9 0 0.4 0.8 0.4 0.4 2 15 111 8 91 16 13 4 26 109 4 71 131 15 2 8 24 12 2 1 1 1 0 5 1
7 (n ¼ 4) 68.8 0.7 11.9 5.1 0 1.3 5.2 0.5 1.7 11 67 91 11 29 19 74 15 66 200 26 316 288 37 8 18 67 29 6 1 1 3 0 10 3
d.s. 0.7 0.1 0.5 0.2 0 0.2 1.2 0.1 0.3 1 9 10 1 3 2 30 6 13 65 5 61 37 1 2 3 6 5 1 1 0 1 0 1 1
8 (n ¼ 1) 62.6 0.7 12.1 4.7 0.1 1.4 9.7 0.5 1.3 11 108 97 11 25 14 37 7 39 274 21 186 343 35 5 16 68 28 5 1 1 2 0 8 2
9 (n ¼ 2) 47.4 0.8 21.7 6.5 0.2 0.8 2.5 1.3 2.9 10 88 48 13 27 29 114 30 174 394 40 384 695 99 8 71 179 70 12 3 2 4 1 32 5
d.s. 2.4 0 0.1 0.1 0 0 0.1 0.1 0.2 1 1 11 2 6 7 19 1 37 25 4 21 208 12 0 28 0 4 1 1 1 0 0 6 1
10 (n ¼ 1) 50.1 0.9 17.0 7.1 0.2 2.8 6.2 1.1 2.7 18 90 141 19 81 57 144 13 124 240 27 160 485 43 4 24 87 38 7 1 1 3 1 13 2
11 (n ¼ 1) 50 1.1 18.3 9.1 0.3 2.9 2.1 1.3 1.9 21 121 177 42 87 54 160 13 83 170 33 238 954 58 6 43 111 51 10 2 1 4 1 15 2
12 (n ¼ 4) 55.3 0.9 15.4 6.4 0.1 2.0 7.0 0.8 2.0 15 88 123 16 53 32 111 14 72 247 29 240 388 42 6 25 79 36 7 2 1 3 0 12 3
d.s. 0.8 0.1 2.4 1.0 0.1 0.8 2.2 0.4 0.5 3 13 28 10 22 14 27 5 25 51 2 58 210 8 2 8 18 6 1 1 0 0 0 3 1
13 (n ¼ 2) 48.3 0.6 11.1 5.3 0.1 1.5 15.2 0.6 1.4 13 75 159 17 77 42 80 18 48 409 23 175 338 32 5 27 62 29 5 1 1 3 0 9 3
d.s. 2.0 0 0.9 0.8 0.1 0.8 1.0 0.3 0.6 2 20 119 7 64 24 17 5 28 86 7 25 33 8 1 6 10 9 1 0 0 1 0 3 1
14 (n ¼ 1) 52.6 0.6 10.4 4.4 0 0.9 13.8 0.3 0.9 11 48 82 12 28 20 70 12 31 288 19 245 277 28 5 27 54 24 5 1 1 2 0 7 2
15 (n ¼ 5) 49.5 0.8 15.3 6.0 0.2 2.4 11.7 1.0 1.9 16 90 125 18 53 42 91 11 71 465 27 140 445 39 3 20 70 31 6 1 1 3 0 11 3
d.s. 1.5 0.1 0.8 0.3 0.1 0.2 1.1 0.2 0.3 1 13 15 4 4 9 7 1 12 134 3 19 113 6 1 8 14 5 0 0 0 1 0 1 1
16 (n ¼ 1) 54.6 0.8 16.0 6.5 0.1 3.4 13.5 0.9 2.3 18 123 153 16 75 41 103 4 70 357 25 140 371 36 4 11 69 32 5 1 0 3 0 12 2
Table 4
Matrix and skeleton description of the nine mineralogicalepetrographic groups (MPG)
MPG CG Matrix Skelethon
Grain size Composition Sintering Orientation Minerals and glass Lithic fragments Microfauna
(in abundance order)
1 1e2 Sandyeclayey Ferric Slight High Qtz Pl Kfs Ms Bt Alm Hbl Ep ss Phyllites e
2 3 Sandyeclayey Ferric Slight High Qtz Pl Kfs Bt Chl Alm Hbl Aug Ep ss. Phyllites e
3 4 Sandy Ferric Slight e Anc Aen Aeg Fa Glass Usp Qtz e e
4 5e6 Clayeyesandy Ferric Slight e Qtz Pl Kfs Ms Bt Chl Alm Hbl Aug Ep ss. Phyllites e
5 PA 15, 29 Sandy Ferric High e Ads Sa Krs Qtz Bt Aug Ol Sps-Alm Mag Glass Volcanic e
6 9 Clayey Ferric Medium Glass Sa Sl Pl Kfs Bt Grs Ol Qtz

E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

e e e
7 7 Sandy Carbonate Slight e Qtz Kfs Ab e e
8 12 Sandyeclayey Carbonate Slight e Qtz Kfs Ab Ms Bt Chl Arenite Rare
9 13e15 Clayey Carbonate Medium e Qtz Kfs Pl Ms Bt Chl Alm Hbl Cal Dol Aug Metamorphic Rare
The correspondence with chemical groups (CG) is indicated in the second column.
Abbreviations: Ab, albite; Ads, andesine; Aeg, aegirine; Aen, aenigmatite; Alm, almandine; Anc, anorthoclase; Bt, biotite; Cal, calcite; Chl, chlorite; Dol, dolomite; Ep, epidote; Grs, grossular; Hbl, hornblende;
Kfs, K-feldspar; Krs, kaersutite; Mg, magnetite; Ms, muscovite; Ol, olivine; Pl, plagioclase; Qtz, quartz; Sa, sanidine; Sl, salite; Sps, spessartine; Usp, ulvöspinel.

Table 5
Micro-chemical analyses of the matrices; mean values (n ¼ 10)
CG 1 CG 2 CG 3 CG 4 CG 5 CG 6 CG 7 CG 9 CG 12 CG 13 CG 15 CG 16
/PA 54 7 8 40 1 29 16 24 20 15 32 3 5 52 53 28 27 35 34 39 46 49 38
SiO2 59.8 60.0 59.9 57.8 53.7 58.0 55.6 60.7 58.5 55.8 56.3 56.4 60.2 51.5 45.4 48.5 55.9 57.2 61.4 45.7 53.5 52.6 55.83
TiO2 1.0 0.9 1.0 1.0 0.8 0.9 1.1 1.0 0.9 1.4 1.0 1.3 1.2 0.9 0.4 0.5 0.9 0.8 0.9 0.7 0.7 1.1 0.905
Cr2O3 0.2 0.3 0.2 0.2 0.2 0.2 0.2 0.4 0.1 0.2 0.2 0.3 0.0 0.2 0.2 0.2 0.4 0.3 0.4 0.3 0.0 0.2 0.37
Al2O3 24.1 22.9 23.5 25.2 26.3 23.9 26.7 22.7 23.7 24.8 26.3 24.8 24.8 26.0 26.4 22.5 25.7 21.5 15.8 17.4 18.6 20.9 16.53
FeO 6.4 7.3 6.8 7.4 9.1 7.5 8.7 6.9 7.8 10.5 7.9 8.1 6.5 12.8 7.4 8.3 6.7 6.4 5.4 7.6 5.9 6.5 5.42
MnO 0.4 0.2 0.3 0.2 0.4 0.1 0.3 0.2 0.2 0.4 0.2 0.2 0.0 0.2 0.2 0.1 1.3 0.4 0.2 0.2 0.1 0.2 0.31
MgO 2.6 2.3 2.4 3.7 3.4 2.2 1.9 2.7 1.7 2.0 2.2 2.3 1.5 3.1 3.2 3.6 1.4 3.2 3.6 2.7 3.4 3.4 3.69
CaO 1.0 2.1 1.6 1.0 1.7 1.8 2.4 2.0 2.2 2.4 0.5 2.9 2.4 2.8 14.8 9.3 3.2 6.8 9.4 22.0 14.6 12.6 13.36
Na2O 1.1 1.4 1.3 1.1 1.4 2.1 1.2 1.0 2.9 1.0 1.5 0.8 1.5 0.6 0.8 1.5 1.6 1.2 1.4 0.9 1.6 0.8 1.285
K2 O 3.4 2.6 3.0 2.5 2.9 3.3 1.9 2.4 2.1 1.5 3.9 2.9 1.8 2.0 1.3 5.6 2.8 2.4 1.7 2.5 1.7 1.7 2.2
Tot. 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
Samples listed based on cluster groupings (CG, chemical group).

1083
1084 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

Table 6
EMP analysis: representative analyses of aenigmatite and glass of MPG 3 samples; kaersutite, biotite, olivine, spessartineealmandine and glass of MPG 5 samples;
sanidine, salite, bytownite and glass (pumice) of MPG 6 samples
1 2 3 4 5 6 7 8 9 10 11 12
Sample PA 19 PA 19 PA 15 PA 29 PA 29 PA 15 PA 15 PA 29 PA 27 PA 27 PA 27 PA 27
Phase/glass Aenigmatite Glass Kaersutite Kaersutite Biotite Olivine Almandinee Glass Sanidine Salite bytownite Glass
spessartine
SiO2 41.40 72.55 40.36 40.60 34.68 33.56 36.29 52.92 65.71 50.50 47.18 62.49
TiO2 9.42 0.41 4.04 4.60 2.23 0.25 0.10 1.22 0.11 0.68 0.01 0.43
Al2O3 0.63 7.73 13.68 13.30 19.60 2.10 19.89 19.08 19.07 2.55 33.30 18.56
FeO 40.77 8.19 11.07 10.40 22.98 36.04 22.46 6.80 0.22 9.61 0.70 2.84
MnO 1.12 0.34 0.11 0.15 0.33 0.54 19.24 0.26 0.00 0.93 0.01 0.21
MgO 0.46 0.02 13.61 13.77 6.28 22.31 0.25 3.49 0.00 11.95 0.07 0.41
CaO 0.57 0.40 12.08 12.21 0.08 0.82 0.31 7.19 0.77 22.78 17.38 1.81
Na2O 6.91 4.11 2.49 2.48 0.12 0.01 0.00 5.03 3.75 0.66 1.51 5.18
K2O 0.00 4.04 0.67 0.69 8.85 0.03 0.00 1.57 10.43 0.01 0.38 7.18
Cl2O 0.02 1.11 0.03 0.00 0.05 0.02 0.00 0.40 0.02 0.04 0.00 0.87
P2O5 0.00 0.02 0.03 0.02 0.01 0.90 0.00 0.60 0.00 0.20 0.22 0.04
SO3 0.00 0.06 0.00 0.04 0.05 0.02 0.00 0.11 0.00 0.00 0.06 0.07
Tot. 101.30 98.98 98.16 98.24 95.26 96.60 98.54 98.67 100.08 99.89 100.80 100.09

Si 5.867 e 5.935 5.951 1.953 0.987 6.062 e 2.977 1.916 2.166 e

Ti 1.004 e 0.447 0.507 0.095 0.005 0.013 e 0.004 0.019 0.000 e
Al 0.106 e 2.371 2.298 1.301 0.073 3.915 e 1.018 0.114 1.802 e
Fe 4.832 e 1.361 1.275 1.082 0.887 3.138 e 0.008 0.305 0.027 e
Mn 0.134 e 0.014 0.019 0.016 0.013 2.722 e 0.000 0.029 0.000 e
Mg 0.097 e 2.984 3.010 0.527 0.978 0.061 e 0.000 0.676 0.005 e
Ca 0.087 e 1.903 1.917 0.005 0.026 0.056 e 0.037 0.926 0.855 e
Na 1.900 e 0.709 0.704 0.013 0.001 0.001 e 0.329 0.048 0.134 e
K 0.000 e 0.126 0.128 0.636 0.001 0.000 e 0.603 0.000 0.022 e
Cl 0.003 e 0.006 0.000 0.004 0.001 0.000 e 0.001 0.002 0.000 e
P 0.000 e 0.000 0.000 0.000 0.000 0.000 e 0.000 0.000 0.000 e
S 0.000 e 0.000 0.004 0.002 0.000 0.000 e 0.000 0.000 0.002 e
Ox. 20 e 23 23 22 4 24 e 8 6 8 e

The generally poorly-preserved white coatings are currently plagioclase, K-feldspar, biotite, chlorite, almandine, horn-
under detailed investigation, while preliminary observations blende, augite and epidote ss. Muscovite is apparently absent
can be provided for the red coatings. The latter show coating in samples from mineralogical group 2, chlorite and augite
layers of variable thickness (from 20 to 50 mm, with an aver- from 1. The three groups are also characterised by the constant
age of 25 mm), characterised by numerous Fe oxide particles presence of phyllitic lithic fragments (sometimes having
dispersed in a scarcely vitrified matrix (Fig. 4). This last char- a coarser grain size, maybe phyllitic schists).
acteristic is observable also in the intermediate layer which The second provenance group includes samples from MPG
measures 100 mm and separates the red coating from the ce- 9, clustered into CG 13 and 15. The mineralogicalepetro-
ramic body proper. Matrix compositions of the intermediate graphic characteristics of samples reflect those of lithologies
layer and of the coating are reported in Table 7. There is typically cropping out in the Calabria area, including Paola;
a gradual increase in Al, Fe, Ti and Mg (compensated by a de- however, the carbonate composition of the matrix (CaO >
crease in Si and K) from the ceramic body proper to the inter- 10.5 wt% in group 15; CaO > 14.5 wt% in group 13) is appar-
mediate layer and the coating. ently incompatible with the composition of sediments here
analysed (CaO < 9 wt%). In this case artifacts produced
6. Discussion from carbonate clays were likely traded on a regional scale.
Archaeometric studies indicate that the ceramics from Locri
Integrating data on both ceramic specimens and sediment (Mirti et al., 2004) are incompatible with samples from
samples with geological data on the area of Paola, 7 different MPG 9; however, based on its geology, the Crati Valley is
provenance groups of ceramics can be established. a possible area of production. Comparisons with the region
The first provenance group includes samples from MPG 1, 2 of Sicily provide no further insight: the absence of volcanic
and 4, clustered into CG 1e3 and 5e6 (except for samples PA 8, lithic fragments excludes the Etna area (cfr. e.g. the products
15 and 29). The chemical and mineralogical composition of this from Lentini and Syracuse in Barone et al., 2005), CaO con-
large group of finds is compatible with that of sediments crop- tents (10e16 wt%) are not comparable to those of products
ping out in the Paola area. CaO contents are modest, the matrices from Agrigento (Alaimo et al., 1997), and from Messina (Bar-
are ferric, and mineral assemblages comprise quartz, one et al., 2005), and the chemical and mineralepetrographic
 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089 1085

Fig. 3. SEMeBSE images: aenigmatite crystal (A) and glass (B) in mineroepetrographic group (MPG) 3 samples; kaersutite crystal (C) and glass (D) in samples
PA 29 and 15 of MPG 5; salite crystal (E) and pumice (F) of MPG 5 samples.

composition is not similar to that of products from Segesta produced at San Giusto (Foggia, Puglia) (Gliozzo et al.,
(see Montana et al., 2003). It is difficult to draw comparisons 2005) but the mineralogy and petrography differ.
with ceramic products from Gela (Sicily) and Sibari (Cala- Although raw materials could have been imported to be
bria) (Barone et al., 2004). The bulk chemical composition worked locally, technological considerations exclude this hy-
is perfectly comparable to that of some cooking ware pothesis (see below).

Fig. 4. SEMeBSE images of the red coating. (A) Ceramic body, intermediate layer and external coating are clearly distinguishable. (B) Zoom on both intermediate
layer and external coating.
1086 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

Table 7 at Berenice (Riley, 1979) and Carthage (Fulford and Peacock,

The mean chemical composition (square SEMeEDS analysis, n ¼ 10) of the 1984).
red coating
The fourth provenance group includes samples from MPG
PA 54 PA 40 3, cluster into CG 4. The bulk chemistry and mineralogical as-
Matrix Intermediate Coating Matrix Intermediate Coating sociation agree perfectly with those reported in the literature
layer layer for ‘‘Pantellerian Ware’’ (Alaimo and Montana, 2003; Fabbri
SiO2 59.8 50.9 47.8 58.2 55.2 46.2 et al., 2004). Also the composition of glass within the ceramic
TiO2 1.0 1.7 1.4 1.0 0.9 1.7 body is similar to that of Pantellerian Ware.
Cr2O3 0.2 0.1 0.2 0.2 0.1 0.3
Al2O3 25.1 28.5 30.2 24.9 26.8 31.3
The fifth provenance group includes samples PA 15 (CG 5)
FeO 6.4 10.9 12.6 7.1 9.3 11.8 and PA 29 (CG 3), from MPG 5. Except for quartz, the mineral
MnO 0.4 0.1 0.1 0.2 0.1 0.3 association in these two samples is similar to that in trachytes
MgO 1.7 2.6 2.7 3.7 3.9 4.3 from Etna consisting of ‘‘plagioclase (An40), augitic clinopyr-
CaO 1.0 1.4 1.2 1.0 0.9 2.0 oxene, FeeTi oxides, some olivine and kaersutite set in
Na2O 1.1 0.7 1.1 1.2 0.8 0.7
K2O 3.4 3.1 2.7 2.6 1.9 1.3
a groundmass containing alkali-feldspar (anorthoclase to Na-
sanidine), clinopyroxene, Ti-magnetite, biotite and sporadic
Tot. 100 100 100 100 100 100
sodalite and nepheline’’ (Peccerillo, 2005, pp. 221e222).
However, it is difficult to justify the presence of almandinee
spessartine (Alm53Sp46Py1, negligible amount of grossular),
The third provenance group includes samples from MPG 6, although this composition is compatible with that of igneous
clustered into CG 9. The SieCa-poor, AleFeeMg rich chem- rocks. The composition of glass within the ceramic bodies
ical composition of the ceramic body is mainly determined by may be compared to that of lava in numerous other areas;
the ferric matrix and by the mineralogical composition charac- we report the composition of lava erupted on the southern
terised by the extreme scarcity of quartz and the abundance of slope of Etna as an example (Cristofolini et al., 1991). The
volcanic glass (pumice). The mineralogical association is typ- known reference groups in Sicily (Barone et al., 2005) are
ical of potassic alkaline volcanism. Comparison between the not ideal, in the sense that the samples are from classes of ce-
composition of pumice fragments enclosed in the ceramic ramics which are rather different, but the comparison reveals
body and that of pumice and volcanic rocks (Peccerillo, a vague similarity between sample PA 29 and products from
2005) excludes the regions of Tuscany and Sardinia. Sicily Catania (‘unvarnished’ and ‘black varnished’ pottery of the
(Etna, Hyblei, Linosa, Pantelleria, Sicilian Channel and Us- Hellenistic period) and between sample PA 15 and products
tica) is another unlikely source area, especially because out- from Messina (‘calcidian’, ‘banded’ pottery and ‘black var-
cropping rocks are generally much poorer in K. The same nished plates’). We also compared the composition of this
consideration holds for the Vulture area, where rocks are gen- group with that of amphorae produced in Lipari with a coarse
erally rich in Ca and Si-poor and Al-rich with respect to the fabric and containing obsidian (Cipriano et al., 2000); even in
studied glass fragments. Lazio is a possible area of prove- this case there is only a partial similarity. Lastly, note that the
nance, the area of the Vulsini (the Alban and Sabatini hills olivine in rocks from Etna is mainly forsteritic (Peccerillo,
are silica- and/or alkali-poor), though Campania offers a wider 2005), whereas that in the studied ceramic specimens is Fe-
spectrum of possibilities and a greater chance for direct com- rich.
parisons. For example, the composition of pumice from Ischia The sixth provenance group includes samples from MPG 1,
(see Ischia Phase I pumice compositions in Poli et al., 1987) clustered into CG 1. The mineralogical and petrographic char-
and from the Flegrean fields (see pumice composition in acteristics of sample PA 8 are comparable to those of artifacts
Orsi et al., 1995; Civetta et al., 1997) is perfectly comparable. produced at Paola. Nevertheless, the archaeological study of
There is less affinity with the composition of the pumice and the sample and its Ni and Cr contents (bulk chemical analysis)
lava of the Somma-Vesuvius and Pontine Island complexes suggest that it came from Greece. As has been demonstrated
(Peccerillo, 2005 with references therein). The mineral assem- (Farnsworth et al., 1977; Jones, 1986; Barone et al., 2002),
blage is compatible with the mineralogy of the Campania area: Cr and Ni contents greater than 200e300 ppm are typical of
abundant bytownite, salitic clinopyroxene, biotite, forsteritic Greek products, while western products have Cr and Ni con-
olivine and Fe oxides are the main constituents of latites and tents of less than 100 ppm. The Fe and Mg contents, which
trachytes. The salite composition is not as aluminous as that are sometimes diagnostic for Greek products, in this case pro-
reported for the Vesuvius area and the northern part of the Ro- vide no further insight. Furthermore, note that the typological
man volcanic province (Deer et al., 1978). The absence of and macroscopic analysis of mixtures suggested that this sam-
feldspathoids is also noteworthy. ple was possibly produced in the Aegean.
The typology of artifacts does not suggest any particular The seventh provenance group includes samples from MPG
area of production, since types 39 (imitation African forms) 8, clustered into CG 12. These samples have a carbonate ma-
and 79 are widespread throughout the Mediterranean. In Italy trix, inclusions typical of North African products (rounded
they have been found at Cosa (Dyson, 1976), Ostia, Carmi- quartz, quartzites), and a bulk chemical composition compara-
niello (Carsana, 1994), Postacrusta (Leone, 2000) and Agri- ble to that of some Tunisian products (personal unpublished
gento (Bonacasa Carra, 1995). Type 79 has also been found data). The presence of a white engobe in PA 47 and 48 recalls
 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089 1087

a traditional North African method of production but provides morphological similarity to the stewpots of Pompeii (PA 1)
no conclusive evidence of provenance, given that the use of and San Foca (PA 13). The imitation of the ‘‘vaso a listello’’,
clay slips was also widespread in southern Italy. widespread in Italy during the Late Roman age, was also pro-
Lastly, for a small group of samples (PA 28 and 53 from duced at Paola (PA 8), where it was also imported (PA 12).
CG 7 and PA 38 from CG 16) it has not been possible to sug- The characteristic ‘casserole’ pans come from Pantelleria,
gest a convincing provenance hypothesis. Uncertainties are while bowls (tableware) and pans (cookware) may come
due to the absence of diagnostic minerals. Sample PA 38 from Etna (Sicily). A stewpot with a morphology similar to
(chemical group 16) stands out because, although its bulk com- that of stewpots from southern Italy (e.g. Carminiello, Ordona,
position is similar to that of characterising chemical groups Agrigento) and North Africa (Carthage, Berenice) comes from
13e15, the high degree of vitrification does not allow the com- Greek territory. Pitchers and basins with forms widespread in
plete mineralogical characterisation of the mixture. the Mediterranean come from North Africa.
Results depict a scenario that is in keeping with that at var-
7. Conclusions ious sites in central-southern Italy for the Late Antique period;
however, the examined assemblage should no doubt be much
The results achieved add significantly to our understanding more extensive in order to provide a complete picture of im-
of the nature of the ceramic production and trade in the late ports and exports.
Antique Mediterranean area, in the sense that they provide From a technological perspective, this study contributes to
the first reference group available for the coarse wares pro- the investigation of an extremely complex scenario requiring
duced at Paola, new data on the ceramic production of Cala- more specific, in-depth research. Ceramic ware from Paola
bria, previously very poorly known, and the reconstruction was produced from a sandyeclayey blend of materials col-
of the commercial trade routes involving Paola settlement at lected in the immediate vicinity of the archaeological site
a Mediterranean scale. and, considering the roundness of some crystals, along river-
The integration of historicalearchaeological data with ar- banks (likely of River San Francesco).
chaeometric data (Fig. 5) revealed that the coarse wares found Different, coarse raw materials were used without prepara-
at the archaeological site of Paola can be ascribed to the local pro- tion. The blending of coarse materials is appropriate for the
duction (first provenance group), to a wider territory correspond- production of cookware, but hardly suitable for the production
ing to the whole regional extension of Calabria (second of fine tableware. The use of cookware mixtures for the pro-
provenance group), to the Campania region (third provenance duction of tableware has often been ascribed to the ‘‘multi-
group) and to Pantelleria (fourth provenance group). The ce- functionality’’ of forms. The example of Paola provides
ramics of the fifth to seventh provenance groups are instead likely evidence for a further possible interpretation: ‘‘inappropriate’’
to be imported from Sicily, Greece and Africa, respectively. mixtures may have been used simply because there were no
Basins, stewpots, pitchers, bowls, pans, casseroles and lids other raw materials in the vicinity and not in a conscious at-
were produced locally, whereas tableware and food storage tempt to create ‘‘multifunctional’’ objects. In other words,
vessels made from calcareous mixtures were imported on a re- the easiest explanation seems to be that the local availability
gional scale. of raw materials dictated the poor-quality technical solution
The imitation African forms (e.g. PA 12 and 53) were no which was nevertheless acceptable from a functional and eco-
doubt produced locally, and it is interesting to note the nomic point of view.

Fig. 5. Schematic map with the results obtained by this study in terms of imports.
1088 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089

The method of preparation and the application of the red D’Ambrosio, B., Mannoni, T., Sfrecola, S., 1986. Stato delle ricerche miner-
coating (a rough, uneven layer) also testify to the coarse alogiche sulle ceramiche mediterranee. In: La ceramica medievale nel
Mediterraneo Occidentale (Siena-Faenza 1984), Firenze, pp. 601e609.
methods of production. Lastly, the low degree of sintering of D’Ambrosio, B., Mannoni, T., Sfrecola, S., 1989. La provenienza delle anfore
the matrix suggests that the ceramics produced at Paola romane di alcuni contesti italiani: possibilità e limiti del metodo mineralo-
were generally fired at intermediate (about 800  C) to low gicoAmphores romaines et histoire économique: dix ans de recherche.
(around 650e750  C) temperatures. Actes du Colloque de Sienne (22e24 mai 1986). Coll. de l’Ecole Française
de Rome 114, 269e284.
D’Andria, F., 1980. San Foca (Lecce). Scavo di un impianto costiero di età
References romano-imperiale. Studi di Antichità 2, 79e88.
Deer, W.A., Howie, R.A., Zussman, J., 1978. Rock-Forming Minerals. 2A.
Alaimo, R., Montana, G., Giarrusso, R., Di Franco, L., Bonacasa Single-Chain Silicates. John Wiley & Sons, New York, N.Y..
Carra, R.M., Denaro, M., Belvedere, O., Burgio, A., Rizzo, M.S., Di Giuseppe, H., 1998. La fornace di Calle (Tricarico): produzione e diffu-
1997. Le ceramiche comuni di Agrigento, Segesta e Termini Imerese: sione. In: Saguı̀, L. (Ed.), Ceramica in Italia: VIeVII secolo, Atti del Con-
risultati archeometrici e problemi archeologici. In: Santoro Bianchi, S., vegno in onore di John W. Hayes (Roma 11e13 maggio 1995), Firenze,
Fabbri, B. (Eds.), Atti della 1 Giornata di archeometria della Ceramica pp. 735e752.
(Bologna 28 febbraio 1997). Il contributo delle analisi archeometriche Dyson, S.L., 1976. Cosa: the utilitarian pottery. Memoirs of the American
allo studio delle ceramiche grezze e comuni. Bologna University Press, Academy in Rome 33.
pp. 46e69. Fabbri, B., Gualtieri, S., Guiducci, G., 2004. Archeometria e prove tecniche
Alaimo, R., Montana, G., 2003. Scienza e archeologia: le analisi archeome- per la ricostruzione di un ciclo produttivo di una ceramica da cucina: il
triche. Archeologia Subacquea e ceramiche da fuoco a Pantelleria. In: San- caso Pantellerian Ware. In: Fabbri, B. (Ed.), Atti della VI Giornata di
toro Bianchi, S., Guiducci, G., Tusa, S. (Eds.), Pantellerian Ware. Dario Archeometria della Ceramica (Ferrara, 9 April 2002). Metodologie di ri-
Flaccovio Editore, Palermo, pp. 52e55. cerca e obiettivi degli studi: lo stato dell’arte. University Press, Bologna,
Annese, C., 2000. Le ceramiche tardoantiche della domus B. In: Volpe, G., pp. 79e87.
Ordona, X. (Eds.). Edipuglia, Bari, pp. 285e342. Farnsworth, M., Perlman, I., Asaro, F., 1977. Corinth and Corfu: a neutron
Annese, C., De Felice, G., Turchiano, M., 2000. Ceramiche della prima e me- activation study of their pottery. American Journal of Archaeology 81,
dia età imperiale dai riempimenti delle latrine della domus A. In: 455e468.
Volpe, G., Ordona, X. (Eds.). Edipuglia, Bari, pp. 251e265. Fracchia, E., 1994. La ceramica comune e da cucina di età repubblicana e della
Arthur, P., 1994. Ceramica comune tardo-antica e alto-medievale. In: Arthur, P. prima e media età imperiale. In: Arthur, P. (Ed.), Il complesso archeologico
(Ed.), Il complesso archeologico di Carminiello ai Mannesi, Napoli (scavi di Carminiello ai Mannesi, Napoli (scavi 1983e1984). Galatina, pp. 173e
1983e1984). Congedo Editore, Galatina, pp. 181e220. 179.
Barone, G., Ioppolo, S., Majolino, D., Migliardo, P., Spagnolo, G., 2002. A Fulford, M.G., Peacock, D.P.S., 1984. The coarse (kitchen and domestic) and
multidisciplinary investigation on archaeological excavation in Messina painted wares. In: Fulford, M.G., Peacock, D.P.S. (Eds.), Excavations at
(Sicily). Part II. A study of the transport amphorae. Journal of Cultural Carthage: The British Mission. The avenue du Président Habib Bourguiba,
Heritage 3, 171e176. Salambo: The pottery and other ceramic objects from the site, Sheffield
Barone, G., Crupi, V., Galli, S., Longo, F., Majolino, D., Mazzoleni, P., vol. I (2), pp. 155e231.
Spagnolo, G., 2004. Archaeometric analyses on ‘‘Corinthian B’’ transport Gliozzo, E., Memmi, I., 2004. Black gloss pottery: production sites and tech-
amphorae found in Gela (Sicily, Italy). Archaeometry 46, 553e568. nology in Northern Etruria. Part I: provenance studies. Archaeometry 46,
Barone, G., Lo Giudice, A., Mazzoleni, P., Pezzino, A., Barilaro, D., Crupi, V., 201e225.
Triscari, M., 2005. Chemical characterization and statistical multivariate Gliozzo, E., Fortina, C., Turbanti Memmi, I., Turchiano, M., Volpe, G., 2005.
analysis of ancient pottery from Messina, Catania, Lentini and Siracusa Cooking and painted ware from San Giusto (Lucera, Foggia): the produc-
(Sicily). Archaeometry 47, 745e762. tion cycle, from the supply of raw materials to the commercialisation of
Bonacasa Carra, R.M., 1995. Agrigento. La necropoli paleocristiana sub divo. products. Archaeometry 47, 13e29.
Roma. Hayes, J.W., 1972. Late Roman pottery. A Catalogue of Roman Fine Wares,
Bonardi, G., Perrone, V., Zappetta, A., 1974. I rapporti tra ‘‘metabasiti’’, ‘‘fil- London.
ladi’’ e ‘‘scisti micacei’’ nell’area tra Paola e Rose (Calabria). Bollettino Henderson, G., Page, B.N.G. (Eds.), 1963. Carta geologica della Calabria e
della Società Geologica Italiana 94, 245e276. Nota illustrativa delle tavolette appartenenti al Foglio 229 ‘‘Paola’’ della
Carsana, V., 1994. Ceramica da cucina tardo-antica ed alto-medievale. In: carta topografica d’Italia (1:100.000), Roma.
Arthur, P. (Ed.), Il complesso archeologico di Carminiello ai Mannesi, Na- Jones, R.E., 1986. Greek and Cypriot pottery. A Review of Scientific Studies,
poli (scavi 1983e1984). Congedo Editore, Galatina, pp. 221e258. Occasional Paper 1. Fitch Laboratory, The British School at Athens,
Chiaromonte Trerè, C., 1984. Ceramica grezza e depurata. In: Bonghi Jovino, Athens.
M. (Ed.), Ricerche a Pompei: L’insula 5 della regio VI dalle origini al 79 Lanzafame, G., Zuffa, G.G., 1976. Geologia e petrografia del Foglio Bisignano
D.C., Roma, pp. 140e193. (Bacino del Crati, Calabria). Geologica Romana 15, 223e270.
Cipriano, S., De Vecchi, G.-P., Mazzocchin, S., 2000. Anfore ad impasto Lattanzi, E., 1999. La Calabria. In: L’Italia meridionale in età tardoantica, Atti
grezzo con ossidiana a Padova: tipologia, impasti, provenienza. In: Brogio- del XXXVIII Convegno di Studi sulla Magna Grecia (Taranto, 2e6 ottobre
lo, G.P., Olcese, G. (Eds.), Convegno Internazionale di Desenzano del 1998), Taranto, pp. 733e756.
Garda (8e10 April 1999). Produzione ceramica in area padana tra il II se- Leone, D., 2000. In: Volpe, G., Ordona, X. (Eds.), Le ceramiche tardoantiche
colo a.C. e il VII secolo d.C.: nuovi dati e prospettive di ricerca. SAP e della fattoria di Posta Crusta. Edipuglia, Bari, pp. 387e432.
Società Archeologica Padana S.r.l., Mantova, pp. 191e197. Mirti, P., Gulmini, M., Pace, M., Elia, D., 2004. The provenance of red figure
Civetta, L., Orsi, G., Pappalardo, L., Fisher, R.V., Heiken, G., Ort, M., 1997. vases from Locri Epizephiri (southern Italy): new evidence by chemical
Geochemical zoning, mingling, eruptive dynamics and depositional pro- analysis. Archaeometry 46, 183e200.
cesses e the Campanian Ignimbrite, Campi Flegrei caldera, Italy. Journal Montana, G., Mommsen, H., Iliopoulos, I., Schwedt, A., Denaro, M., 2003.
of Volcanology and Geothermal Research 75, 183e219. The petrography and chemistry of thin-walled ware from an Ellenistice
Cotton, M.A., 1979. The Late Republican Villa at Posto Francolise. British Roman site at Segesta (Sicily). Archaeometry 45, 375e389.
School at Rome, London. Orsi, G., Civetta, L., D’Antonio, M., Di Girolamo, P., Piochi, M., 1995. Step-
Cristofolini, R., Corsaro, R.A., Ferlito, C., 1991. Variazioni petrochimiche filling and development of a three-layers magma chamber: the Neapolitan
nella successione etnea: un riesame in base a nuovi dati da campioni di Yellow Tuff case history. Journal of Volcanology and Geothermal Research
superficie e da sondaggi. Acta Vulcanologica 1, 25e37. 67, 291e312.
 E. Gliozzo et al. / Journal of Archaeological Science 35 (2008) 1074e1089 1089

Ostia III ¼ AA. VV., 1973. Ostia III. Le terme del nuotatore, Roma. Sangineto, A.B., 2000. Trasformazioni o crisi nei Bruttii fra il II a. C. ed il VII
Peacock, D.P.S., 1982. Carthage and Cossyra: a ceramic conundrum. In: d. C.? In: Lo Cascio, E., Storchi Marino, A. (Eds.), Modalità insediative e
Actes colloque sur la céramique antique (Carthage 1980), Carthage, strutture agrarie nell’Italia meridionale in età romana, Atti del Convegno di
pp. 91e98. Napoli (11e13 giugno 1998), Bari, pp. 203e246.
Peacock, D.P.S., 1984. Petrology and origins. In: Fulford, M.G., Santoro Bianchi, S., Guiducci, G., Tusa, S. (Eds.), 2003. Pantellerian Ware.
Peacock, D.P.S. (Eds.), Excavations at Carthage: The British Mission. Archeologia Subacquea e ceramiche da fuoco a Pantelleria. Dario Flacco-
The avenue du Président Habib Bourguiba, Salambo: The Pottery and vio Editore, Palermo.
Other Ceramic Objects from the Site, Sheffield vol. I (2), pp. 6e20. Semeraro, G., 1992. La ceramica comune preromana e romana. In:
Peccerillo, A., 2005. Plio-Quaternary Volcanism in Italy. Springer. D’Andria, F., Whitehouse, D. (Eds.), Excavations at Otranto. The Finds,
Piccarreta, G., 1973. Rapporti tra le metamorfici affioranti nella zona com- vol. II. Congedo Editore, Galatina, pp. 63e78.
presa tra Castiglione, Nicastro, Conflenti e la basse valle del fiume Savuto Small, A.M., Buck, R.J., 1994. The excavations of San Giovanni di Ruoti. I.
(Calabria). Bollettino della Società Geologica Italiana 92, 423e433. The Villas and Their Enviroment, Toronto-Buffalo-London.
Poli, S., Chiesa, S., Gillot, P.-Y., Gregnanin, A., Guichard, F., 1987. Chemistry Turchiano, M., 2000. La cisterna e il suo contesto. Materiali tardoantichi dalla
versus time in the volcanic complex of Ischia (Gulf of Naples, Italy): ev- domus B. In: Volpe, G., Ordona, X. (Eds.). Edipuglia, Bari, pp. 343e385.
idence of successive magmatic cycles. Contribution to Mineralogy and Pe- Vivacqua, P., 2002. La ceramica di uso comune dello scavo di Paola Stadio
trology 95, 322e335. (Cosenza). Degree thesis-University of Siena.
Riley, J.A., 1979. The coarse pottery from Benghazi. In: Lloyd, J.A. (Ed.), Ex- Whitehouse, D., Costantini, L., Guidobaldi, F., Passi, S., Pensabene, P.,
cavations at Sidi Khrebish, Benghazi (Berenice) II. Supplements to Libya Pratt, S., Reece, R., Reese, D., 1985. The Schola Praeconum II. Papers
Antiqua, vol. 5, pp. 91e497. of the British School at Rome 53, 163e210.