Roman pottery production in the coast of Granada (Spain) in the early Late Antiquity: a scientific analysis of ceramics from the kiln site of Los Matagallares

doi:10.21203/rs.3.rs-4696732/v1

Download PDF

Research Article

Roman pottery production in the coast of Granada (Spain) in the early Late Antiquity: a scientific analysis of ceramics from the kiln site of Los Matagallares

https://doi.org/10.21203/rs.3.rs-4696732/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The kiln site of Los Matagallares (Salobreña, Granada, Spain) is one of the most important pottery production sites of the Roman period in southern Iberia, particularly for the 3rd century AD. The workshop produced a wide variety of ceramics, but focused on the manufacture of amphorae for the long-distance trade of local/regional fish products, wine, and other foodstuffs, and thus played an important role in the economy of Roman Baetica during the transition to the Late Roman Empire. A combined petrographic, mineralogical, and elemental characterisation of amphorae and other ceramics from the site, using a multi-technique approach (thin-section petrography, X-ray diffraction and wavelength dispersive X-ray fluorescence), allowed for establishing a compositional reference group for this workshop, as well as investigating technological aspects of pottery production at the site. The definition of a compositional reference group for the particular kiln site enabled us to identify an amphora from Los Matagallares in a consumption context in the Catalan coastal city of Mataró. More generally, it will be an invaluable tool for future studies, as it will allow for the identification of products from this prolific kiln site in consumption contexts all over the Roman world, and will aid towards a better understanding of the trade connections between production areas and consumption sites in Antiquity.

Roman pottery

Amphorae

Hispania

ceramic petrography

WD-XRF

ceramic technology

The coast of the Roman province of Hispania Baetica (roughly today’s Andalusia in southern Spain), played an important economic role in Antiquity, thanks ―among others― to the trade of fish sauces and salted fish, as well as, occasionally, of wine and other foodstuffs. Closely linked to the food production industries, a large number of pottery workshops arose along the coastline of the Strait of Gibraltar, between southern Iberia and northern Morocco, which were producing amphorae to commercialise foodstuffs (see Bernal-Casasola and Lagóstena 2004; Bernal-Casasola 2006; Díaz 2014; Díaz and Bernal-Casasola 2017). During the Early Roman Empire, the main hub of this trade was Gades (Cádiz), but in the 3rd century AD a large part of production moved to regions such as Lusitania, Cartagena, and the Mediterranean coast of Andalusia between Málaga and Granada. In the case of the Granada coastal area, various pottery kiln sites are documented in the zone of Salobreña, in the surroundings of the Roman Selambina, next to the mouth of the Guadalfeo river (Bernal-Casasola and Navas 1998; Fernández-García 2004; Bernal-Casasola 2006). The most prominent of these workshops, for its production capacity and the variety of vessels produced, was that of Los Matagallares (Fig. 1) (Bernal-Casasola 1998a; Bernal-Casasola et al. 2004a). Due to the disparity between the number of kiln sites identified in this zone and the scarce number of known food production centres (fish factories and maritime villae), it has been proposed that this was a specialised pottery production zone that was intended to meet the demand for amphora containers from both urban and rural food production centres in the wider region (Bernal-Casasola 1999, 2006, 2023; Bernal-Casasola et al. 2004a). This would have included ―among others― the fish factories of Sexi (Almuñécar), the main Roman centre in the coast of Granada, whose products were likely oriented towards a long-distance maritime trade (Sotomayor 1971; Molina and Jiménez 1983; Ponsich 1988, 177; Lagóstena 2001, 153–154).

Considering the importance that this pottery production area must have held for the Baetican economy in the 3rd century AD and later, the characterisation of the output of its most prolific workshop is crucial, in order to better understand the trade from the coast of Granada in Roman times. In this study we present a compositional and technological characterization of amphorae and other ceramics manufactured in the kiln site of Los Matagallares, based on their combined petrographic, mineralogical, and elemental analysis. A first scientific study on some amphorae from this workshop (Vigil de la Villa et al. 1998a) had started to shed light on the composition of local ceramic products, but a series of important issues remained unexplored. Based upon that initial study, the present research aims to achieve two main objectives. On the one hand, it seeks to establish a compositional reference group (or groups) for this workshop, which would enable future studies to identify products from Matagallares in consumption sites all over the Roman Empire. This in turn will help to better understand not only the magnitude, but also directions of the trade of foodstuffs in amphorae from the coast of Granada in Roman times. On the other hand, the scientific analysis of the various types of amphorae and other locally manufactured ceramic materials will also enable a better understanding of the pottery production technology at this particular Roman workshop. Linking technological aspects (e.g. paste recipes) and ceramic forms/types, potentially allows insights into the organisation of production. Technological issues have only rarely been considered by previous studies on Roman pottery workshops, not only those located in our study area but more generally along the entire coast of Roman Baetica, with the exception of a small number of studies that have started to progressively fill this gap (e.g. García Vargas and Lavado 1995, Barrio and García Giménez 1998; Vigil de la Villa et al. 1998a, 1998b; Martínez et al. 1999; Gómez Morón et al. 2001; Domínguez-Bella et al. 2004).

THE ROMAN POTTERY WORKSHOP OF LOS MATAGALLARES

The kiln site of Los Matagallares, located 3 km northwest of the Roman city of Selambina, present-day Salobreña, is one of the best-known Roman pottery workshops on the Mediterranean coast of Andalusia (N36°46′ W3°36′). Archaeological excavations conducted between 1995 and 1999 (Bernal-Casasola and Navas 1996, 1999; Bernal-Casasola 1998a; Bernal-Casasola et al. 2004a) enabled a good understanding of the organization and chronology of the workshop, as well as of the range of its products. During these excavations, the remains of up to five kilns, a series of waste dumps and other structures linked to the activity of the workshop were uncovered (Fig. 1b-d). The material findings also included a variety of ceramic potter’s tools, including smoothers, rolls, and bitroncoconical supports ―likely used during the forming and drying of vessels―, as well as fragmentary balls of fired clay (Lorenzo and Bernal-Casasola 1998). The stratigraphic sequence and the material context revealed that the workshop was active from the end of the 2nd or early 3rd century AD to the late 3rd or early 4th century AD. Its importance lies in being one of the best examples of pottery kiln sites in southern Iberia in the 3rd century AD, a key transitional period between the Early and Late Roman Empire.

A wide variety of amphora types were manufactured at Los Matagallares (Bernal-Casasola 1998b; Fernández-García 2004). Most were intended to transport fish products (i.e. garum and other fish sauces, as well as salted fish), especially the types Dressel 14 and Almagro 51c, which appear to have been the most commonly produced here. For each of these two types, up to four morphological variants (labelled A to D in each case) were differentiated. Other amphora types for fish products that were manufactured at this site, but have been recovered in lesser amounts, were Beltrán IIB, Keay XVI, Keay XLI, and Beltrán 72. Despite the apparent predominance of vessels related to the fish trade, amphorae likely used for wine were also manufactured, notably the types Dressel 30, imitations of Gauloise 4, and Matagallares I are frequently attested, in addition to types such as Matagallares II and Baelo I which are less commonly found. Amphorae for the transport of oil, particularly a late variant of the type Dressel 20, were also produced in the workshop, but were only recovered in very low quantities (Bernal-Casasola 1998b).

Despite a clear focus on the production of amphorae for food trade, other ceramics were also manufactured in this workshop, including a wide range of common and cooking wares, which consisted of local forms as well as imitations of North African casseroles and lids (Bernal-Casasola 1998c; Bernal-Casasola et al. 1998a, b), but also building materials (Torrecilla 1998). The commercial distribution of these other wares was likely more restricted than that of the amphorae.

A set of 44 ceramics from the workshop were sampled for analysis (Table 1; Fig. 2 ―for illustrations of the entire analysed assemblage, see Online Resource 1―). Most come from stratigraphic units (SU) related to the use, abandonment, and collapse of kilns C, D and ―in a few cases― B, excavated during the 1999 field season (Bernal-Casasola et al. 1999, 2004a). To date, hardly any of the portable materials from this latest excavation campaign at the site had been published, unlike previous campaigns (see Bernal-Casasola 1998a, c); this study is thus the first time this ceramic assemblage is presented. Only one of the samples analysed comes from the 1995 excavation at the site (Bernal-Casasola 1998a), and is not directly related to a particular kiln structure but was retrieved from a waste dump (Table 1).

The selected samples are mostly of amphorae, including different variants of types Dressel 14 (n = 12) and Almagro 51c (n = 7), as well as a few samples of types Dressel 30 (n = 4), Matagallares I (n = 4), Keay XVI (n = 3), and one each of Keay XLI and Dressel 20 parva. Additionally, common wares (n = 7), cooking wares (n = 2), building materials (n = 2), and one ceramic bitroncoconical support used as a potter’s tool were also subjected to analysis (Table 1; Fig. 2; Online Resource 1). Since the aims were to establish compositional reference groups and assess technological choices, the sampling strategy was designed so that results would allow to explore the compositional and fabric variability among the different ceramic classes and types manufactured in the workshop, but also within each of the most frequent amphora types (i.e. Dressel 14, Almagro 51c, or Dressel 30).

After a first macroscopic examination with the aid of a binocular microscope, all the ceramic samples were thin sectioned for petrographic analysis by optical microscopy (OM). The thin sections were examined using an Olympus BX41 polarising microscope, working with magnifications between ×20 and ×200. The analysis and description of ceramic fabrics were undertaken following an adjusted version of the systems proposed by Whitbread (1989, 1995) and Quinn (2013).

In a second stage, a sub-set of 20 samples ―including 16 amphorae, 2 cooking wares, and 2 common wares (Table 1)― were further examined through wavelength dispersive X-ray fluorescence (WD-XRF) spectroscopy for their elemental characterisation and through powder X-ray diffraction (XRD) for gaining additional information on their mineralogical composition. For both analyses, a sample of each ceramic was powdered and homogenised ―after removing the external surface― in a tungsten carbide mill and dried at 120ºC for at least 24 h. The WD-XRF analysis was conducted at the Fitch Laboratory of the British School at Athens, while XRD was carried out at the Centres Científics i Tecnològics-Universitat de Barcelona (CCiTUB).

The elemental analysis was performed using a Bruker S8 TIGER 4kW WD-XRF spectrometer with Rh excitation source. Samples were measured as glass beads prepared from 1g of ignited sample and 6g of a mixture of lithium metaborate/lithium tetraborate with lithium bromide added as non-wetting agent. Twenty-six major, minor and trace elements were determined (Na, Mg, Al, Si, P, K, Ca, Ti, Fe, V, Cr, Mn, Co, Ni, Cu, Zn, Rb, Sr, Y, Zr, Ba, La, Ce, Nd, Pb, and Th) using a custom calibration based on 43 certified reference materials (Georgakopoulou et al. 2017). While determined with the particular calibration routine, Co is not reported and it was not considered in the evaluation of results in this study due to contamination during sample preparation.1

XRD measurements were performed using a PANalytical X’Pert PRO MPD alpha 1 diffractometer, working with Cu–Kα radiation (l = 1.5406 Å). Spectra were taken from 5 to 80° 2θ, using a step-size of 0.026°2θ and a step-time of 47.5 s. The evaluation of crystalline phases was carried out using the software HighScore Plus by PANalytical, which includes the Joint Committee of Powder Diffraction Standards (JCPDS) data bank. The mineralogical assemblage identified through this technique allowed for an estimation of equivalent firing temperatures (EFT) for each sample (e.g. Roberts 1963; Maggetti 1982; Riccardi et al. 1999; Cultrone et al. 2001; Cau 2003; Maritan 2004).

CHARACTERIZATION OF LOCAL FABRICS

Macroscopic characteristics

Macroscopically, the ceramic fabrics in the analysed samples showed similarities in the types of inclusions (Fig. 3–4), which are somewhat heterogeneous, ranging from glossy, colourless to white or light grey (quartz-like) particles, to others which are greyish in colour, and vary from spherical to platy, and from glossy to matt. Variable amounts of fine-grained silver and golden mica can also be seen, as well as few to rare matt white-yellowish (carbonate-like) inclusions. Despite these general similarities, the fabrics are rather variable in terms of the colour of the matrix. Most of the samples showed various hues of brown or reddish brown, usually homogeneous, but sometimes with heterogeneities within the same sample, such as bicoloured wall sections with a darkened (greyish) core. Completely dark-coloured sections are very rare. These variations are observed in both the amphorae (Fig. 3) and the cooking/common wares (Fig. 4), and appear unrelated to the ceramic class or type, the only exception being a much more heterogeneous matrix in MTG-37, which is the only sample of tegula (roof tile) analysed (Fig. 4k). It should be noticed that the creamish-whitish colour observed in the surfaces of many samples in this workshop seem related to post-depositional alterations, as the same concretions are commonly found on the breaks of the pottery sherds.

Petrographic-mineralogical composition

Thin section petrography showed a very similar fabric in almost all analysed samples (except for samples MTG-20, a common ware, and MTG-37, a building material), characterised by predominant metamorphic inclusions (Fig. 5–6). However, some variability was observed, especially in terms of textural aspects. Inclusions (> 0.01 mm) are frequent (15–25%, rarely 30%) and show a bimodal grain size distribution, with a moderately abundant coarse fraction formed by medium-coarse sand (0.25-1.00 mm) and few-rare very coarse sand (1.00–2.00 mm). However, about one third of the samples present a finer texture, with lower amounts of medium and, especially, coarse sand. In any case, a gradual transition between these coarse and fine fabrics is observed.

The inclusions in the coarse fraction (0.25-5.00 mm, but generally not coarser than 2.00 mm) consist of predominant metamorphic rock fragments mostly composed of quartz and micas (muscovite and, in lower frequency, biotite or chlorite), likely derived from schists. These fragments can occasionally also contain feldspars, garnet, epidote, and more rarely staurolite, cordierite, or sillimanite. Polycrystalline quartz/quartzite is frequent-common in the coarse fraction, while monocrystalline quartz is rare. Coarse carbonate inclusions are common to rare, and consist mainly of micritic lumps, in addition to very rare sparite and shell fragments. The coarse fraction also comprises few-rare garnet and opaques/iron oxides, and rare alkali feldspar, while only a few samples contain one or more of other accessory mineral inclusions, like staurolite, cordierite, amphibole, clinopyroxene, kyanite, andalusite, and chert. In some samples (especially in MTG-10, MTG-12, and MTG-40), occasional but very large fragments (up to 5.00 mm) of quartzarenite are observed (Fig. 6a); the shape and boundaries of these fragments, as well as the fact that the very fine sandy grains of quartz that they contain are very similar to those in the fine fraction of the fabric, suggest that these may be naturally occurring inclusions of the clay source, rather than added temper.

The fine fraction (0.01–0.25 mm) is very abundant and composed of predominant monocrystalline quartz and very thin micas (mostly muscovite, with mode < 0.05/0.10 mm), in addition to common opaques/iron oxides and polycrystalline quartz, common-rare carbonate inclusions (mostly micrite), as well as very few-rare alkali feldspar, plagioclase, and accessory heavy minerals that may include epidote, amphibole, staurolite, garnet, clinopyroxene, tourmaline, or zircon.

The porosity in this fabric is relatively low (2–5%), with small-sized vughs and vesicles (generally < 0.5 mm); elongated voids are rare, except in one of the cooking ware samples (MTG-17) in which they are frequent.

The colour of the clay matrix usually ranges from brown to orange or reddish brown in PPL, and dark brown to dark orange/reddish brown in XP; less commonly it is light yellowish brown in PPL, and dark yellowish brown in XP. It is generally homogeneous, although variations related to core/margin differentiation ―sometimes with a darkened core― were observed in a few samples. In most samples the matrix is optically active to slightly active, under XP, except in rare cases (MTG-1, MTG-4, MTG-19, MTG-27, and MTG-38) in which the matrix is optically inactive (e.g. Figure 6b, j). The latter also show a higher degree of alteration/decomposition of carbonate and mica inclusions, all together suggesting higher firing temperatures than in the rest of the samples.

The results from XRD mineralogical analysis provided additional support to this interpretation (Table 2). Most of the 20 samples that were analysed through this technique revealed the absence of clear firing phases in a low Ca environment, which only allows us to propose that the EFT was likely not higher than 900/950 ºC; nevertheless, the optical activity of the clay matrix observed in the thin sections of these samples allows us to refine this estimation and suggests that these temperatures may not have been higher than 800/850 ºC (Quinn 2013, 191). The presence of hematite in these samples may be interpreted either as a primary phase or as a firing phase under oxidising conditions, in the latter case pointing to EFTs closer to 900/950 ºC. Small reflections of maghemite in a few samples (Table 2) may indicate firing at an EFT above 750 ºC under reducing conditions (Travé et al. 2019), although this phase has also been reported for ceramics fired at lower temperatures (Maritan 2004). Only two of the samples analysed by XRD, MTG-1 and MTG-4, showed evidence for firing phases typical of calcareous ceramics, such as gehlenite and diopside, in this case pointing to an EFT of 850–950 ºC, whereas the low calcareous sample MTG-19 showed spinel and maghemite and firing phases, which, along with the near absence of phyllosilicates, indicate an EFT of about 950/1000 ºC or higher (Table 2).

Besides the main petrographic fabric, found in 42 of the 44 samples analysed by OM, two fabric variants were identified in samples MTG-20 and MTG-37, both sharing the same general features of the main fabric except for some particularities (Fig. 7). In MTG-20 (Fig. 7a-c), a much more reddish-coloured, iron-rich clay matrix is observed, along with more common staurolite inclusions in the fine fraction; these features may indicate the use of a slightly different clayey sediment compared to the rest of the ceramic assemblage. As for MTG-37 (Fig. 7d-f), which is the only roof tile sample analysed, it is differentiated by a much more heterogeneous matrix, with very prominent clay streaks that range from brown-orange streaks with almost no inclusions, to yellowish streaks with frequent quartz and carbonate inclusions. Although these features may suggest incomplete clay mixing, other hypotheses (e.g. natural inhomogeneities of the raw clayey sediment) cannot be ruled out.

Table 2 Mineralogical composition and equivalent firing temperature (EFT) of the 20 ceramic samples analysed by XRD. Abbreviations for minerals (based on Kretz 1983): Qtz, quartz; Pl, plagioclase; Kfs, K-feldspar; Cal, calcite; Gh, gehlenite; Px, pyroxene; Ill-Ms, illite-muscovite; Hem, hematite; Mgh, maghemite; Spl, spinel

Elemental composition

The elemental data for the 20 ceramic samples that were analysed by WD-XRF (Table 3) reveal clear similarities in the bulk composition of most of the samples. One sample, MTG-20, stands out, due to higher concentrations of Fe₂O₃ and Al₂O₃, and lower of CaO, Sr, MgO, SiO₂, and Mn, than the rest of the data set. Most of the compositional variability in the remaining data set seems related to CaO, which fluctuates between 1.7% and 7.3%. Indeed, also the compositional variation matrix or CVM (Aitchison 1986, 2005; Buxeda 1999), calculated over the entire data set indicates that CaO, with a τ._i value of 6.18, is by far the most variable element (Fig. 8). Other elements that account for more than 50% of the variability are Mn (τ._i = 1.71), Ba (τ._i = 1.35), P₂O₅ (τ._i = 1.32), Sr (τ._i = 1.31), Na₂O (τ._i = 1.23), and Rb (τ._i = 1.23). Some of these, in particular P₂O₅ but also potentially Na₂O and Mn, may be affected by alteration/contamination, while Sr is geochemically related to Ca and reflects variability in the latter. The total variation value (vt) obtained from the CVM is 0.59. This drops to 0.46 when sample MTG-20 is excluded, and to 0.28 when leaving out CaO, reaching a value that falls within the expected range for monogenic populations (Buxeda and Kilikoglou 2003).

Cluster analysis (CA) of the elemental data (Fig. 9a), after excluding the P₂O₅ and Pb to avoid possible contamination problems, revealed the presence of one large cluster including all the samples except MTG-20. Within this cluster, four samples (MTG-1, MTG-4, MTG-5, and MTG-16) form a separate subcluster (Fig. 9a) due to a slightly elevated CaO content (4.5–7.3%), compared to the low calcareous composition that characterises the rest of the assemblage, and also due to a slightly lower content in K₂O in these four samples (Table 3). However, except for these variations, all the samples in the main cluster are very similar in their bulk compositions. This can also be visualised by principal component analysis (PCA), where almost all the samples ―except MTG-20― form a large group with some internal variability mainly related to CaO and, to a lesser degree, Mn (Fig. 9b).

Table 3 Elemental composition of the 20 ceramic samples from the kiln site of Los Matagallares that were analysed by WD-XRF. Concentrations of oxides (and LOI) are given in %, trace elements in ppm

It is worth mentioning that no clear relation was observed between these variations in CaO content as determined by WD-XRF and the frequency of carbonate inclusions observed in thin section, nor with the presence of secondary depositions of calcite, which were not particularly abundant within the ceramic body of any sample. Although coarse and fine carbonate inclusions vary from common to rare in the studied fabrics, not all of the more calcareous samples showed common carbonate inclusions in thin section (e.g. these are rare in MTG-16). Thus, the combined petrographic and elemental data suggest that these variations in CaO may be at least partially related also to the composition of the clay matrix rather than to differences in the abundance of aplastic carbonate inclusions alone. Of course, another possibility would be that these latter were not homogeneously distributed throughout the ceramic body, which could have influenced the CaO content determined by WD-XRF analysis.

In summary, despite some variation in CaO content, it is possible to define a compositional reference group for the ceramic products from the kiln site of Los Matagallares. The average bulk composition of this reference group is shown in Table 4. The sample MTG-20 was excluded from this reference group, as this is a unique form of common ware (a trilobed jar) which varies in both petrographic fabric and elemental composition. For the moment, we cannot ascertain whether this particular vessel was manufactured locally, representing a slightly different recipe, or is the product from another, most likely regional workshop.

MANUFACTURING TECHNOLOGY

The petrographic and elemental analysis of ceramic samples from Los Matagallares not only enabled the characterization of local fabrics and the definition of a compositional reference group, but also provided important data concerning technological aspects linked to their production, including raw material selection and processing, paste preparation, and firing.

First, as regards raw material procurement, the petrographic and mineralogical composition of the ceramic fabrics is compatible with the metamorphic geology that characterises the surroundings of the workshop, supporting the hypothesis formulated by Vigil de la Villa et al. (1998a) that locally available raw materials had been used in production. The terrain where the site is located is dominated by outcrops of the Alpujárride Complex, which mainly consist of metamorphic rocks such as schists, quartzites, and amphibolites, containing in some cases minerals such as staurolite, garnet, kyanite, sillimanite, and andalucite (Aldaya et al. 1980; García-Dueñas and Avidad 1981; Junta de Andalucía 1998) (Fig. 10). The workshop is also located close to water streams like the Rambla de Molvízar and the lower Guadalfeo river, the former flowing into the latter about 3 km southeast from the site. This implies the occurrence of alluvial deposits with contribution of materials from various other sources, including also large outcrops of carbonate sedimentary rocks to the east. In any case, medium-high grade metamorphic outcrops predominate the landscape around the site (García-Dueñas and Avidad 1981).

In a first study on the ceramic products from Los Matagallares, Vigil de la Villa et al. (1998a) reported on the availability of clay sources in the vicinity of the site, which may have been potentially used for pottery production. The authors also highlight the finding of numerous metamorphic pebbles in one of the waste dumps of the workshop, which had apparently been brought from a nearby streamlet; according to the authors, these pebbles could potentially have been crushed to be used as temper (Vigil de la Villa et al. 1998a). However, the shape of the rock fragments observed in thin section often varies between subangular and subrounded in the same sample, instead of being mainly angular as it would be expected in the case of crushed material (Quinn 2013, 165). Moreover, there are variations in the frequency of coarse aplastic inclusions (> 0.5 mm) in the ceramic fabrics, from samples with abundant medium-coarse sand to others where these are much scarcer. For these reasons, the possibility that the coarse rock fragments in the ceramic fabrics were naturally occurring inclusions within the raw clay appears more likely. Knowing the range of compositional variation within raw material sources available in the area may help to further assess this issue. In the present case no differences in elemental composition were found between samples with high or low frequency of coarse inclusions, which would be expected in the case of tempering with a material coming from a different source, provided it has a sufficiently distinct elemental composition to that of the raw clay base (Neff et al. 1988, 1989).

The textural variability observed in the petrographic fabrics may be the result of natural variations of the exploited clay deposits or potentially of variations in the amount of added temper, but also, alternatively, differences in the refining of clays ―for example through sieving or levigation― during raw material processing. Whatever the case, no relation was observed between the coarseness of the clay paste and the function of the vessel, as the same range of variability was found in amphorae, common wares, and cooking wares. However, a possible trend was observed regarding the type of amphora: most ―but not all― of the analysed samples of types Almagro 51c, Dressel 30, and Matagallares I, as well as the only sample of type Keay XLI, showed a finer fabric, while this was present in only one of the 12 samples of type Dressel 14 analysed. The amphorae of this type were generally made in a coarser fabric, as well as the few samples analysed of types Keay XVI and Dressel 20 parva. Even if the relation is not completely linear, this trend may suggest slight variations in the source of raw materials and/or the paste recipes depending on the type to be produced. This correlation does not seem to be related either to the specific function of the amphorae (e.g. vessels used for fish products can be found in both fine and coarse variants) or to a temporal shift (given the overlapping chronological range of many of these amphora types), but most likely to the size of the vessel, since types Dressel 14 and Keay XVI were normally larger than the others, so it is possible that these formal variations may have influenced technological choices regarding the coarseness of clay pastes.

In any case, despite these minor variations, the strong similarities in elemental composition and petrographic fabric observed in most of the ceramic samples from Los Matagallares point to the use of overall similar raw materials and paste preparation techniques. The slight variation in the CaO content found through WD-XRF elemental analysis does not have a clear correlation with any petrographic feature and is more likely related to differences in the composition of the clay matrix rather than in the aplastic inclusions, as discussed above. In this case, however, it was observed that the few samples with slightly elevated calcium content correspond to amphorae, while all samples of cooking and common wares that were chemically analysed were low calcareous. These results may suggest that, even if low calcareous clay pastes were predominantly used for all the ceramic classes, there may be higher inconsistencies in raw material selection, processing and/or paste preparation in the production of amphorae. Analysis on a larger number of samples would be needed, however, to confirm this pattern.

Two of the ceramic samples analysed deserve a special mention. One is sample MTG-20, a trilobed jar, which was manufactured using a slightly different clayey raw material, as revealed by its distinct both petrographic and elemental composition. Since the rock fragments and mineral inclusions present in this sample are still compatible with the local geology, it is difficult to ascertain whether this vessel was produced in the same workshop using another nearby clay source or in another regional workshop. The latter alternative seems plausible considering its particular form compared to the rest of the ceramic assemblage; it must be noted that only very rare examples of trilobed jars have been reported at the site (Bernal-Casasola et al. 1998a, 343). The other sample is MTG-37, a roof tile, which has a very heterogeneous fabric ―as observed in thin section and also to the naked eye― that might be the result of either incomplete clay mixing or the use of a natural variegated clay that was not completely homogenised (see discussion in Ho and Quinn 2021). These inhomogeneities are not uncommon in ancient ceramic building materials, which were often made from poorly processed clays (Quinn 2013, 213). However, this is not always the case, as can be seen in the other sample of building material analysed, MTG-38, which showed the same petrographic fabric as the rest of the assemblage and, therefore, indicates that at Los Matagallares in some cases building materials were manufactured using the same paste recipes as those used for producing amphorae and cooking/common wares.

Another essential step of the chaîne opératoire involved in pottery production concerns the firing process. For the majority of the ceramic samples analysed, the combined OM and XRD analysis point to relatively low firing temperatures, likely not higher than c. 800/850ºC. Evidence for higher firing temperatures was found only in a few amphora samples (MTG-1, MTG-4, MTG-19 and MTG-27) and, interestingly, in the two samples of building materials analysed (MTG-37 and MTG-38). Although the macroscopic fabrics often presented a homogenous brown or reddish brown colour, the occurrence of variations in the hues or colours of the fabrics, including ―in some samples― core-margin colour heterogeneities in a single ceramic body (Fig. 3–4), indicate that the ceramics manufactured at the kiln site were fired under variable reducing-oxidising conditions (Picon 2002). This variability can be found both in amphorae and in common/cooking wares and could suggest low standardisation as regards the colour of the final ceramic products, independently of their intended function. In any case, these observations on colour and firing conditions should be treated carefully, bearing in mind that the analysed ceramic samples were found either in waster dumps or around the kilns, and so may include unsuccessfully fired products that were not traded.

THE CHARACTERIZATION OF LOCAL FABRICS AND ITS IMPLICATIONS FOR THE STUDY OF DISTRIBUTION AND TRADE

Thin section OM revealed that almost all the ceramics analysed from the workshop of Los Matagallares ―including amphorae, cooking and common wares, and some of the building materials― are very similar in petrographic fabric, while WD-XRF analysis allowed for the definition of a compositional reference group, given the strong similarities in elemental composition among the local ceramic products. The characterisation of the local fabrics and the establishment of a reference group is a prerequisite in order to be able to trace the commercial distribution of the ceramics produced at this kiln site, by unambiguously identifying Los Matagallares products when recovered from consumption sites, both regional and overseas. This is particularly important in the case of amphorae, which were intended to trade local/regional foodstuffs by ship. Given the importance of Baetican amphora trade in Roman times, it would not be unexpected that the amphorae from Los Matagallares ―transporting fish, wine, and other products likely from Selambina, Sexi, and other sites in the coast of Granada― could have been marketed across the Mediterranean and Atlantic coasts of Europe and North Africa and, in fact, evidence for this has been found in sites such as Lyon and in different Mediterranean shipwrecks (Bernal-Casasola 2008, 2016) based on the macroscopic examination of the materials. In the absence of epigraphic evidence, which is only very rarely found in amphorae from Los Matagallares, the study of ceramic fabrics becomes essential for the identification of products from this workshop in consumption sites, as we know that similar forms of amphorae were also produced in other workshops. This implies comparing the petrographic and/or elemental composition of ceramics found in consumption sites with the petrographic fabric and compositional reference group defined for the workshop.

An example of this methodological procedure can be given from consumption sites in northeastern Iberia. As a result of an analytical research programme carried out a few years ago on amphorae from various Late Roman contexts in the current Catalan coastal territory ―the eastern part of Hispania Tarraconensis in Roman times―, a set of 37 amphorae presumably from Baetica were analysed petrographically and elementally to gain further details on their provenance (Fantuzzi and Cau 2019a). In that study, the combined scientific and typological information enabled the identification of broad production areas for most of the amphorae, but the absence of compositional reference groups from various Baetican kiln sites prevented more accurate provenance determination. These materials can now be restudied in the light of the new evidence, to check if any of the amphorae found in Catalan sites could come from the workshop of Los Matagallares. Indeed, comparison between the petrographic fabric group presented in this study and the fabrics defined for Baetican amphorae in Catalan consumption sites (Fantuzzi and Cau 2019a) reveals a very good correspondence with one amphora, ILU088, found at the Roman site of Iluro (present-day Mataró) (Fig. 11a). Also, the WD-XRF elemental composition of this amphora (see Fantuzzi 2015; Fantuzzi and Cau 2019a) matches well with the compositional reference group established for Los Matagallares products, except for a very slightly higher content in K₂O and lower in Na₂O (Fig. 11b). In any case, the strong similarities in petrographic fabric leave no doubt of this relation. Sample ILU088 corresponds to an amphora of type Dressel 30 (Cela and Revilla 2004: Fig. 117n. 18), which appears to have been one of the most common forms manufactured in the workshop of Los Matagallares. The context where this amphora was found is dated to the 6th century AD, but it contained various residual materials from the Late Roman Republican period onwards (Cela and Revilla 2004, 252–253), certainly including this amphora.

A distribution map of amphorae from Los Matagallares in consumption sites remains yet to be drawn. Epigraphic evidence has been of little help, as stamped marks are very rare in this workshop; they have been found only on a few examples of Dressel 14 amphorae, always consisting of a stamp IAN applied on one of the handles (Fig. 2) (Bernal-Casasola 1997, 1998b). Most of the findings of Dressel 14 amphorae recovered from the workshop ―more than 97% of the known examples (Bernal-Casasola 1998b, 303)―, as well as all the other amphora types and local ceramics, were found unstamped. Exceptional examples of oil amphorae Dressel 20 found in Rome would have been manufactured in the area of Selambina based on the tituli picti that they bear, possibly in Los Matagallares or in other nearby ceramic workshops such as Los Barreros (Martínez Rodríguez et al. 2017). Apart from these very occasional epigraphic findings, another indicator to track the distribution of amphorae from this workshop would potentially be the morphology of the containers. However, most of the amphora types produced in Los Matagallares were also manufactured in other kiln sites across southern Iberia and other nearby regions (see Bernal-Casasola 2001, 2019; Fabião 2008; García Vargas and Bernal-Casasola 2008). Even the form Matagallares I, which was first defined for this workshop, is now also known to have been produced at the kiln site of El Mojón in Murcia (Berrocal 2012), so that some of the examples of this type found in consumption sites like Lyon and Vienne in France (Lemaître and Bonnet 2001) could come from either of the two production sites. This emphasizes the need for examination of the ceramic fabrics, ideally through petrographic and/or elemental analysis, as an essential tool for comparison between vessels from consumption sites and those recovered from the production site/s, as can be clearly seen from the example of the amphora ILU088 found in Iluro/Mataró, in northeastern Spain.

The scientific analysis of a large set of ceramics from the kiln site of Los Matagallares, which included amphorae of various types (Dressel 14, Almagro 51c, Dressel 30, Matagallares I, Keay XVI, and others), as well as common wares, cooking wares, and building materials, provided new information on the composition and manufacturing technology of these materials. It also contributed to a better understanding of the workshop's amphora production, as it has allowed for the verification of the local production of some types of amphorae that were previously unknown at this figlina, such as the Dressel 20 type in its parva version.

Thin section petrography showed the presence of similar fabrics in almost all the ceramics analysed, and the WD-XRF elemental analysis also revealed very similar compositions that allowed for the definition of a compositional reference group for the products from the kiln site. Slight variations were found regarding the calcareous content of the clay pastes, as low calcareous clays were normally used except for a few amphorae that were made from slightly calcareous clays. Also, some textural variation was observed in the fabrics, with Dressel 14 amphorae generally made in coarser fabrics, while Almagro 51c, Dressel 30, and Matagallares I often ―but not always― show finer-grained textures. Nevertheless, despite these minor variations, the local ceramics from the workshop were overall very similar in their composition and production technology, including the raw materials used, paste processing techniques, and the firing process. The combined results of OM, WD-XRF, and XRD also indicate that the variability observed in macroscopic fabrics among local pottery products, with variations even in amphorae of the same type, are more related to inconsistencies in firing conditions than to compositional differences.

The only samples that did not fit the compositional reference group belong to a form of common ware (MTG-20), for which a provenance in another nearby regional workshop cannot be ruled out, and a roof tile (MTG-37) whose fabric ―with a very heterogeneous matrix― was the result of using a less processed clay paste than the rest of the assemblage, an aspect that is not infrequent in ancient ceramic building materials.

In summary, the reference group that has been established for the workshop of Los Matagallares, with well-defined petrographic fabrics and elemental compositions, will be useful in future studies to identify amphorae and other ceramic products manufactured at this site when found in consumption sites, as has been illustrated with the example of an amphora found in Iluro/Mataró (NE Spain). Further analyses of relevant vessels from consumption sites will certainly help to better understand the range of the commercial distribution of the products from Los Matagallares. This is important not only regarding amphorae but also other ceramic classes, particularly the locally produced imitations of African cooking wares, which may have been mistaken for imports from Africa Proconsularis in consumption sites.

This case study illustrates well the need for studies on the characterization of pottery production sites in Roman Baetica more generally, as a prerequisite to understand the trade connections between production areas and consumption sites in Antiquity. To date only relatively few studies have been published on the compositional characterization of Roman amphora workshops in Baetica (e.g. Vigil de la Villa et al. 1998b; Grubessi and Conti 1999; Gómez Morón et al. 2001; González Vílchez et al. 2001; Polvorinos del Río et al. 2003; Bernal-Casasola et al. 2004b; Corrales et al. 2011; Fantuzzi and Cau 2019b). In this context it is promising that, when comparing the amphorae from Los Matagallares, on the coast of Granada, with those from other production zones of the Mediterranean coast of Andalusia characterised in recent years ―notably the case of the coast of Málaga province (e.g. Fantuzzi and Cau 2017; Fantuzzi et al. 2020)―, there are clear differences in composition and fabric. The possibility of differentiating between amphorae from different production zones is crucial for their successful identification at consumption sites and, by extension, for the study of trade connections in Antiquity. It is hoped that the promising results obtained so far will encourage analogous analytical investigations into other pottery production sites in the region, contributing to our understanding of the regional economy in a period which saw important political and societal changes across the wider Roman Empire.

Competing interests:

The authors declare no competing interests.

Author Contribution

Conceptualisation: [L.F., D.B.C., J.J.D.]; Methodology: [L.F., D.B.C.]; Investigation: [L.F., D.B.C., J.L.P.S.; M.A.C.O.; E.K.; N.S.M.]; Formal analysis and validation: [L.F., M.A.C.O.; E.K.; N.S.M.]; Writing–original draft: [L.F.]; Writing-Review & Editing: [D.B.C.; J.L.P.S.; M.A.C.O.; E.K.; N.S.M.]; Visualisation (preparation of figures): [L.F., J.L.P.S.]; Funding acquisition: [L.F., J.J.D.]; Field study & sample providing: [D.B.C., L.L.M.]; Supervision: [L.F., D.B.C.]. All authors read and approved the final manuscript.

ACKNOWLEDGEMENTS

This paper is an outcome of the connected projects ARQ-ANALYTICs: Comercio maritimo y producciones anfóricas en el Conventus Gaditanus: análisis arqueométrico de tres casos de estudio (Gades, Baelo Claudia y Selambina) (PI: L. Fantuzzi) and ARQ-ANALYTICs: Analíticas y Nuevas Tecnologías aplicadas a la Arqueología de la producción de las sociedades marítimas púnicorromanas en la bahía de Cádiz (PI: J. J. Díaz), both funded by Fundación CEI·MAR (Proyectos de Jóvenes Investigadores CEI·MAR 2019, ref. CEIJ-C03.1 and CEIJ-C03.2). This is also part of the activities of the Equip de Recerca Arqueològica i Arqueomètrica de la Universitat de Barcelona (ERAAUB), funded by AGAUR of the DIUE of the Generalitat de Catalunya (2021 SGR 00696). It is also an outcome of the projects GARVM IV (PID2022-138814OB-I00/AEI) funded by the Plan Estatal de I + D + I of the Spanish Government/FEDER and MEDUSA (PCM_00031), of the so called Plan Complementario de Ciencias Marinas of the Junta de Andalucía.

Data Availability

Data is provided within the manuscript.

Aitchison J (1986) The statistical analysis of compositional data. Chapman and Hall, London
Aitchison J (2005) A concise guide to compositional data analysis. In: 2nd Compositional Data Analysis Workshop (CoDaWork’05). Universitat de Girona, Girona. http://ima.udg.edu/Activitats/CoDaWork05/A_concise_guide_to_compositional_data_analysis.pdf. Accessed 1 April 2024
Aldaya F, Vera JA, Fontboté JM (1980) Mapa Geológico de España, E. 1:200.000, Granada-Málaga, Hoja 83. Instituto Geológico y Minero de España, Madrid
Barrio J, García Giménez R (1998) Estudio tecnológico y arqueométrico de las producciones cerámicas del taller: análisis de los defectos térmicos de cocción. In: Bernal-Casasola D (ed) Excavaciones arqueológicas en el alfar de la Venta del Carmen. Los Barrios (Cádiz). Ayuntamiento de Los Barrios. Los Barrios / Ediciones UAM, Madrid, pp 277–289
Bernal-Casasola D (1997) Novedades de epigrafía anfórica en la Bética. Talleres costeros granadinos del Bajo Imperio. Bol Asoc Esp Amigos Arqueol 37:99–110
Bernal-Casasola D (ed) (1998a) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña
Bernal-Casasola D (1998b) Las producciones anfóricas del taller. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 231–305
Bernal-Casasola D (1998c) La producción de cerámicas africanas de cocina en la Bética. Aportaciones del taller de Los Matagallares y de la alfarería de Los Barreros (Salobreña, Granada). L’Africa Romana 12:1341–1353
Bernal-Casasola D (1999) Transporte local/regional de envases vacíos en época romana. A propósito de dos talleres anfóricos de época altoimperial (El Rinconcillo, Algeciras, Cádiz) y del Bajo Imperio (Los Matagallares, Salobreña, Granada). II Congreso de Arqueología Peninsular (Zamora 1996), Tomo IV, Arqueología romana y medieval. Fundación Rei Afonso Henriques, Zamora, pp 359–363
Bernal-Casasola D (2001) La producción de ánforas en la Bética en el s. III y durante el bajo imperio romano. Congreso Internacional Ex Baetica Amphorae. Conservas, aceite y vino de la Bética en el Imperio Romano (Écija, Sevilla, 1998). Gráficas Sol, Écija, pp 239–372
Bernal-Casasola D (2006) La industria conservera romana en el Círculo del Estrecho. Consideraciones sobre la geografía de la producción. L’Africa Romana 16:1351–1394
Bernal-Casasola D (2008) Ánforas y vino en la Antigüedad Tardía. El ejemplo de la Hispania meridional. In: Blanquez J, Celestino S (eds) Simposio Internacional Arqueología del vino. El vino en época tardoantigua y medieval. Universidad Autónoma de Madrid, Madrid, pp 33–60
Bernal-Casasola D (2016) Matagallares I (Costa Bética). In: Amphorae ex Hispania. Paisajes de producción y de consumo. Institut Català d'Arqueologia Clàssica, Tarragona. http://amphorae.icac.cat/amphora/matagallares-i-baetica-coast. Accessed 1 June 2024
Bernal-Casasola D (2019) Ánforas tardorromanas en Hispania. Claves de identificación. In: Fernández Ochoa C, Morillo Cerdán A, Zarzalejos Prieto M (eds) Manual de cerámica romana IV. Producciones cerámicas de época medio-imperial y tardorromana. Comunidad de Madrid, Madrid, pp 549–670
Bernal-Casasola D (2023) Puertos y retrotierras haliéuticos en el Fretum Gaditanum. El modelo microrregional de las ánforas vacías revisitado. In: Pasquinucci M, Facella MJ (eds) Porti antichi e retroterra produttivi. Strutture, rotte, merci. Pisa University, Pisa, pp 435–456
Bernal-Casasola D, García Giménez R, Sáez Romero AM, Díaz JJ, Montero R (2004b) Ánforas locales e importadas del yacimiento Los Cargaderos. Caracterización arqueométrica de algunos talleres alfareros de época romana de San Fernando (Cádiz). In: Feliu MJ, Martín J, Edreira MC, Fernández MC, Martínez MP, Gil A, Alcántara R (eds) Avances en arqueometría 2003. Servicio de Publicaciones de la Universidad de Cádiz, Cádiz, pp 145–154
Bernal-Casasola D, Lagóstena L (eds) (2004) Figlinae Baetica. Talleres alfereros y producciones cerámicas en la Bética romana (ss. II a.C-VII d.C). Actas del Congreso Internacional (Cádiz. John and Erica Hedges Ltd., Oxford. 12–14 de noviembre de 2003
Bernal-Casasola D, Lorenzo L, Navas J (1999) Informe preliminar de la intervención arqueológica de urgencia en el yacimiento romano de Los Matagallares (Salobreña, Granada). Unpublished excavation report
Bernal-Casasola D, Lorenzo L, Navas J (2004a) La tipología de los hornos béticos en el s. III d.C. Novedades del taller de los Matagallares (Salobreña, Granada). In: Bernal-Casasola D, Lagóstena L (eds) Figlinae Baeticae. Talleres, alfareros y producciones cerámicas en la Bética romana (ss. II a.C.-VII d.C.) actas del Congreso Internacional (Cádiz. John and Erica Hedges Ltd., Oxford, pp 489–506, pp 12–14. de noviembre de 2003
Bernal-Casasola D, Navas J (1996) Los Matagallares. Un centro alfarero romano en Salobreña. Rev Arqueol 182:42–51
Bernal-Casasola D, Navas J (1998) La producción alfarera en la costa granadina en época romana. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 63–100
Bernal-Casasola D, Navas J (1999) Informe sobre la intervención arqueológica de urgencia en los alfares romanos de Los Matagallares (Salobreña, Granada). Anuario Arqueológico de Andalucía '95, III, Actividades de Urgencia. Consejería de Cultura de la Junta de Andalucía, Sevilla, pp 200–208
Bernal-Casasola D, Navas J, Lorenzo L, Gómez E (1998a) Las cerámicas comunes de producción local. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 307–362
Bernal-Casasola D, Navas J, Lorenzo L, Gómez E (1998b) Las imitaciones de africanas de cocina. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 363–394
Berrocal MC (2012) Producciones anfóricas en la costa meridional de Carthago-Spartaria. In: Bernal-Casasola D, Ribera A (eds) Cerámicas hispanorromanas II. Producciones regionales. Servicio de Publicaciones de la Universidad de Cádiz, Cádiz, pp 255–277
Buxeda J (1999) Alteration and contamination of archaeological ceramics: the perturbation problem. J Archaeol Sci 26:295–313
Buxeda J, Kilikoglou V (2003) Total variation as a measure of variability in chemical data sets. In: Van Zelst L (ed) Patterns and process: a Festschrift in honor of Dr. Edward V. Sayre. Smithsonian Center for Materials Research and Education, Washington DC, pp 185–198
Cau MA (2003) Cerámica tardorromana de cocina de las Islas Baleares: Estudio arqueométrico. Archaeo, Oxford
Cela X, Revilla V (2004) La transició del municipium d’Iluro a Alarona (Mataró). Cultura material i transformacions d’un espia urbà entre els segles V i VII dC. Laietània 15, Matar&#243
Corrales P, Compaña JM, Corrales M, Suárez J (2011) Salsamenta malacitano. Avances de un proyecto de investigación. Itálica, Revista de Arqueología Clásica de Andalucía 1: 29–49
Cultrone G, Rodriguez-Navarro C, Sebastian E, Cazalla O, De la Torre MJ (2001) Carbonate and silicate phase reactions during ceramic firing. Eur J Mineral 13:621–634
Díaz JJ (2014) Los alfares romanos en Hispania (s. II a.C. - VII d.C.). Sistematización de la documentación del conventus Gaditanus y análisis comparativo interprovincial. PhD Dissertation, Universidad de Cádiz
Díaz JJ, Bernal-Casasola D (2017) De las alfarerías de Baetica: focos de producción, tecnología y análisis microespacial de las estructuras de producción. In: Fabião C, Raposo J, Guerra A, Silva F (eds) Olaria Romana: seminário internacional e ateliê de Arqueologia experimental. UNIARQ-Centro de Arqueologia da Universidade de Lisboa, Lisboa, pp 139–174
Domínguez-Bella S, Sánchez A, Domínguez JC, Sánchez M (2004) La producción cerámica en la Bahía de Cádiz en época púnica y romana: análisis arqueométrico de las materias primas minerales y las producciones anfóricas. In: Martín J, Feliu MJ, Edreira MC (eds) Avances en arqueometría 2003. Universidad de Cádiz, Cádiz, pp 155–160
Fabião C (2008) Las ánforas de Lusitania. In: Bernal-Casasola D, Ribera A (eds) Cerámicas hispanorromanas: un estado de la cuestión. Servicio de Publicaciones de la Universidad de Cádiz, Cádiz, pp 725–745
Fantuzzi L (2015) Caracterización arqueométrica de ánforas tardorromanas del nordeste de la península ibérica. PhD Dissertation, Universitat de Barcelona
Fantuzzi L, Cau MA (2017) Investigating the provenance of the Baetican amphorae Dressel 23: new archaeometric evidence from Late Roman consumption centres. Mediterr Archaeol Archaeom 17(1):47–68. https://doi.org/10.5281/zenodo.258085
Fantuzzi L, Cau MA (2019a) Estudio arqueométrico de ánforas béticas procedentes de contextos tardorromanos del nordeste de la península ibérica. In: Coll J (ed) Opera fictiles. Estudios transversales sobre cerámicas antiguas de la península ibérica. Actas del IV Congreso Intenacional de la SECAH-Ex Officina Hispana. La Ergástula, Madrid, pp 201–216
Fantuzzi L, Cau MA (2019b) Amphora production in the Guadalquivir valley (Spain) during the Late Roman period: petrographic, mineralogical, and chemical characterization of reference groups. Archaeol Anthropol Sci 11(12):6785–6802. https://doi.org/10.1007/s12520-019-00937-x
Fantuzzi L, Kiriatzi E, Sáez AM, Müller NS, Williams IICK (2020) Punic amphorae found at Corinth: Provenance analysis and implications for the study of long distance salt-fish trade in the Classical period. Archaeol Anthropol Sci 12:179. https://doi.org/10.1007/s12520-020-01093-3
Fernández-García MI (2004) Alfares y producciones cerámicas en la provincia de Granada. Balance y perspectivas. In: Bernal-Casasola D, Lagóstena L (eds) Figlinae Baeticae. Talleres, alfareros y producciones cerámicas en la Bética romana (ss. II a.C.-VII d.C.) actas del Congreso Internacional (Cádiz. John and Erica Hedges Ltd., Oxford, pp 195–238, pp 12–14. de noviembre de 2003
García-Dueñas V, Avidad J (1981) Mapa Geológico de España. E. 1:50.000. Hoja 1055: Motril. Instituto Geológico y Minero de España, Madrid
García Vargas E, Bernal D (2008) Ánforas de la Bética. In: Bernal D, Ribera A (eds) Cerámicas hispanorromanas: un estado de la cuestión. Servicio de Publicaciones de la Universidad de Cádiz, Cádiz, pp 661–687
García Vargas E, Lavado ML (1995) Ánforas alto, medio y bajoimperiales producidas en el alfar de Puente Melchor (= Villanueva, Paso a Nivel: Puerto Real, Cádiz). Spal 4:215–228
Georgakopoulou M, Hein A, Müller NS, Kiriatzi E (2017) Development and calibration of a WDXRF routine applied to provenance studies on archaeological ceramics. X-Ray Spectrom 46:186–199. https://doi.org/10.1002/xrs.2745
Gómez Morón A (2001) Polvorinos del Río Á, Fernández S Caracterización arqueométrica de las producciones del alfar de El Rinconcillo (Cádiz, España). In: Gómez BM, Respaldiza MA, Pardo ML (eds) III Congreso Nacional de Arqueometría (Sevilla, 1999). Universidad de Sevilla, Sevilla, pp 239–250
González Vílchez MC, González Rodríguez M, Aitlahsen Y (2001) Estudio arqueométrico de asas selladas de ánforas Dressel 20 del yacimiento arqueológico de La Catria (Lora del Río, Sevilla). In: Blázquez JM, Remesal J (eds) Estudios sobre el Monte Testaccio (Roma) II. Publicacions Universitat de Barcelona, Barcelona, pp 401–420
Grubessi O, Conti L (1999) Progetto Testaccio. Studio archeometrico delle anfore Dressel 20: nota II. Confronto analitico tra i reperti spagnoli Tejarillo e La Catria e M. Testaccio. In: Blázquez JM, Remesal J (eds) Estudios sobre el Monte Testaccio (Roma) I. Publicacions Universitat de Barcelona, Barcelona, pp 513–542
Junta de Andalucía (1998) Mapa Geológico-Minero de Andalucía. Escala 1:400.000 (3ª Versión). Dirección General de Industria, Energía y Minas, Sevilla
Kretz R (1983) Symbols for rock-forming minerals. Am Mineral 68:277–279
Lagóstena L (2001) La producción de salsas y conservas de pescado en la Hispania romana (II a.C.-VI d.C). Universidad de Barcelona, Barcelona
Lemaître S, Bonnet C (2001) Des amphores de type Matagallares I importées à Lyon? Congreso Internacional Ex Baetica Amphorae. Conservas, aceite y vino de la Bética en el Imperio romano (Écija y Sevilla 1998). Gráficas Sol, Écija, pp 817–824
Lorenzo L, Bernal-Casasola D (1998) Los útiles de alfarero. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 431–449
Maggetti M (1982) Phase analysis and its significance for technology and origin. In: Olin JS, Franklin AD (eds) Archaeological Ceramics. Smithsonian Institution, Washington DC, pp 121–133
Maritan L (2004) Archaeometric study of Etruscan-Padan type pottery from the Veneto region: Petrographic, mineralogical and geochemical-physical characterisation. Eur J Mineral 16:297–307. https://doi.org/10.1127/0935-1221/2004/0016-0297
Martínez J, Alonso C, García R (1999) La producción cerámica en la provincia de Cádiz. Fabricación de ánforas y materiales complementarios. In: Bendala M, Rico C, Roldán L (eds) El ladrillo y sus derivados en la época romana. Casa de Velázquez / Universidad Autónoma de Madrid, Madrid, pp 45–63
Martínez Rodríguez F, Márquez Cruz M, Sarompas Cazorla CE (2017) Tituli picti en ánforas olearias béticas alusivos a la statio sel y a la statio sex, hallados en el monte Testaccio de Roma, y su contexto geo-histórico. Cuad Prehist Arqueol Univ Granada 27:373–397
Molina F, Jiménez S (1983) La factoría de salazones El Majuelo. In: Molina F (ed) Almuñécar, arqueología e Historia. Caja Provincial de Ahorros de Granada, Granada, pp 279–290
Neff H, Bishop RL, Sayre EV (1988) A simulation approach to the problem of tempering in compositional studies of archaeological ceramics. J Archaeol Sci 15:159–172
Neff H, Bishop RL, Sayre EV (1989) More observations on the problem of tempering in compositional studies of archaeological ceramics. J Archaeol Sci 16:57–69
Picon M (2002) Les modes de cuisson, les pâtes et les vernis de la Graufesenque: une mise au point. In: Genin M, Vernhet A (eds) Céramiques de la Graufesanque et autres productions d’époque romaine. Nouvelles recherches. Hommages à Bettina Hoffmann. Monique Mergoil, Montagnac, pp 139–163
del Polvorinos AJ, Flores V, Tabales MA, Hernández MJ (2003) Caracterización y tecnología de materiales cerámicos romanos de los ss. I a III d.c procedentes del Hospital de las Cinco Llagas de Sevilla. Bol Soc Esp Cerám Vidr 42(2):93–99
Ponsich M (1988) Aceite de oliva y salazones de pescado. Factores geo-económicos de Bética y Tingitania. Universidad Complutense de Madrid, Madrid
Quinn PS (2013) Ceramic petrography. The interpretation of archeological pottery & related artefacts in thin section. Archaeo, Oxford
Riccardi MP, Messiga B, Duminuco P (1999) An approach to the dynamics of clay firing. Appl Clay Sci 15:393–409
Roberts JP (1963) Determination of the firing temperature of ancient ceramics by measurement of thermal expansion. Archaeometry 6:21–25
Sotomayor M (1971) Nueva factoría de salazones de pescado de Almuñécar (Granada). Noticiario Arqueol Hisp 15:145–178
Torrecilla A (1998) Los materiales constructivos. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 395–429
Travé E, Quinn P, Álvaro K (2019) Another one bites the dust: quality control and firing technology in the production of medieval greyware ceramics in Catalonia, Spain. Archaeometry 61:1280–1295. https://doi.org/10.1111/arcm.12495
de la Vigil R, Cuevas J, García R (1998b) Caracterización mineralógica y físico-química de las pastas de las cerámicas de producción local. In: Bernal-Casasola D (ed) Excavaciones arqueológicas en el alfar de la Venta del Carmen. Los Barrios (Cádiz). Ayuntamiento de Los Barrios. Los Barrios / Ediciones UAM, Madrid, pp 291–299
de la Vigil R, Cuevas J, García R, Bernal-Casasola D (1998a) Analítica mineralógica y físico-química de las cerámicas de Los Matagallares y Los Barreros. In: Bernal-Casasola D (ed) Los Matagallares (Salobreña, Granada). Un centro romano de producción alfarera en el siglo III d.C. Primeros resultados de las excavaciones arqueológicas de las campañas de 1995 y 1996. Ayuntamiento de Salobreña, Salobreña, pp 471–496
Whitbread IK (1989) A proposal for the systematic description of thin sections towards the study of ancient ceramic technology. In: Maniatis Y (ed) Archaeometry: proceedings of the 25th international symposium. Elsevier, Amsterdam, pp 127–138
Whitbread IK (1995) Appendix 3. The collection, processing and interpretation of petrographic data. Greek transport amphorae: A petrological and archaeological study. British School at Athens, Athens, pp 365–396

Tables 1 to 4 are available in the Supplementary Files section.

Samples had been milled in tungsten carbide cell, which results in contamination of Co, reflected in the unusually high (and variable) Co contents observed in the entire data set.

No competing interests reported.

Table1.xls
Table2.xls
Table3.xlsx
Table4.xls
OnlineResource1.pdf
Online Resource 1 Illustrations of the 44 ceramic samples analysed from the kiln site of Los Matagallares: A- amphorae of type Dressel 14; B- amphorae of type Almagro 51c; C- amphora of type Keay XLI; D- amphorae of type Matagallares I; E- amphorae of type Keay XVI; F- amphorae of type Dressel 30; G- amphora of type Dressel 20 parva; H- common wares; I- cooking wares; J- potter’s tool (bitroncoconical support); K- ceramic building materials (roof tiles)

Download PDF

Editorial decision: Revision requested
20 Aug, 2024
Reviews received at journal
19 Aug, 2024
Reviews received at journal
28 Jul, 2024
Reviewers agreed at journal
18 Jul, 2024
Reviewers agreed at journal
11 Jul, 2024
Reviewers invited by journal
08 Jul, 2024
Editor assigned by journal
08 Jul, 2024
Submission checks completed at journal
07 Jul, 2024
First submitted to journal
06 Jul, 2024

You are reading this latest preprint version

Roman pottery production in the coast of Granada (Spain) in the early Late Antiquity: a scientific analysis of ceramics from the kiln site of Los Matagallares

Status:

Version 1

Abstract

Figures

INTRODUCTION

THE ROMAN POTTERY WORKSHOP OF LOS MATAGALLARES

MATERIALS AND ANALYTICAL METHODS

CHARACTERIZATION OF LOCAL FABRICS

Macroscopic characteristics

Petrographic-mineralogical composition

Elemental composition

MANUFACTURING TECHNOLOGY

THE CHARACTERIZATION OF LOCAL FABRICS AND ITS IMPLICATIONS FOR THE STUDY OF DISTRIBUTION AND TRADE

SUMMARY AND CONCLUSIONS

Declarations

Competing interests:

Author Contribution

ACKNOWLEDGEMENTS

Data Availability

References

Tables

Footnotes

Additional Declarations

Supplementary Files

Status:

Version 1