The study, published in Archaeological and Anthropological Sciences, examined 59 fragments selected from documented production contexts at two sites about 150 metres apart in the city’s historic pottery district. The material dated mainly from the late eleventh to the mid-twelfth centuries, spanning the Taifa, Almoravid and early Almohad periods. One workshop contained a rod kiln that had been reused as a waste dump after abandonment; the other had a clay-setting pit and production debris. The researchers regarded these contexts as more securely documented than many assemblages of decorated medieval pottery found in residential areas.

Each fragment was mounted in resin and examined by optical microscopy, while all 59 were analysed with scanning electron microscopy and energy-dispersive X-ray spectroscopy, or SEM–EDS. Twenty specimens were prepared as 30-micrometre petrographic thin sections, 14 were tested by wavelength-dispersive X-ray fluorescence, and 18 underwent X-ray diffraction. The methods were chosen to connect visible vessel traits with microscopic fabric, elemental composition and mineral phases formed during firing. Because SEM–EDS can give broader readings across heterogeneous pastes, its results were combined with petrography and XRF rather than treated as a complete chemical map.

Petrography separated the assemblage into three principal fabric groups that broadly followed function. Kitchenware belonged to Fabric F1, whose dark, fine-grained matrix contained abundant phyllite, mica schist and granite inclusions. Its cooking pots were generally low in calcium and sodium but high in silicon, with average values of about 1.5 weight per cent CaO, 0.6 per cent Na₂O and 63.2 per cent SiO₂ in the SEM–EDS dataset. Quartz-rich particles and rock fragments would have increased resistance to thermal shock. Most cooking pots were handmade, although three of the analysed examples were wheel-thrown, documenting a technological transition within the same production setting.

Water jugs formed a distinct group, Fabric F2. Their bodies were poorer in silica and richer in calcium and magnesium than the cooking pots, with a CaO/MgO ratio near 1.25. The authors interpreted this composition as evidence that at least two clays were combined: a medium-grained, non-calcareous red material resembling that used for kitchenware and a lighter clay containing dolomitic inclusions. This blended paste was also chemically similar to that used for kiln rods, suggesting that available workshop clays were adapted according to the demands of different objects. The water vessels were wheel-thrown and decorated with painted red or black designs, while their handles were sometimes made by hand.

Tableware used a finer, calcareous fabric divided into two subgroups. Both had relatively low porosity and fewer inclusions than the kitchenware, but F3-S1 contained more potassium while F3-S2 contained more sodium. Their average K₂O concentrations were about 2.6 and 0.8 per cent, respectively, while Na₂O averaged approximately 1.1 and 2.4 per cent. Neither subgroup consistently belonged to one workshop, indicating that the chemical contrast was not a simple division between the two sites. Instead, the researchers proposed that it could reflect different local clay sources, clay mixing or paste-preparation practices. Geological sampling would be required to determine whether the workshops exploited the same deposits.

The glazes showed that shared clays did not imply a single firing recipe. SEM–EDS identified 26 glazed samples, including 11 kitchenware and 15 tableware examples, along with glaze droplets from workshop equipment. Many transparent glazes contained lead and were coloured honey-yellow by iron oxide. Transparent tableware glazes ranged from 27.1 to 56.2 per cent PbO. Comparison with the ceramic bodies indicated that lead compounds were combined with quartz sand rather than applied as a simple lead coating. The tableware glazes also showed evidence of a deliberately added potassium-bearing material, possibly derived from woody plant ash, while some recipes apparently included a sodium-rich component such as salt.

Colour was produced through a limited but technically varied palette. Copper oxide generated green transparent glazes, with some samples also containing zinc, possibly indicating the reuse of brass. Brown decoration on honey-coloured tableware contained iron and manganese oxides and may have been painted onto the biscuit-fired vessel before glazing. Opaque white glaze came from cassiterite crystals formed when tin oxide was added to lead glaze. Turquoise combined copper with tin, the latter dispersing fine cassiterite particles that increased opacity. A droplet adhering to a workshop handle contained partially dissolved quartz, cassiterite and lead glaze, consistent with preparation of tin “calx,” while its wider composition included substances associated with lustre pigments. The evidence demonstrates use of lustre-related materials, but the researchers stressed that it did not conclusively prove lustreware production at either workshop.

Principal component analysis reinforced the separation of the three fabric groups: the first three components accounted for 96 per cent of the total chemical variance, with 76.9 per cent contributed by the first component alone. Mineral phases also suggested varying firing temperatures. Kitchenware containing dehydrated muscovite was probably fired below 850 degrees Celsius, while some waterware and tableware indicated temperatures of about 850 degrees or higher; analcime in one tableware subgroup pointed to vitrification above 950 degrees Celsius. Workshop equipment showed repeated exposure to heat: kiln rods contained abundant glassy phases, and one tripod stilt appeared to have been fired more than once. Overall, the evidence points to a coherent but flexible craft system, with functional choices in clay selection and multiple glazing traditions rather than highly specialised production. The study’s principal limitation is that the two selected workshops and 59 fragments cannot establish the exact source of the clays or represent every medieval pottery tradition in Almería.

Subject of Research: Archaeometric characterization of medieval pottery production, clay fabrics, glazes and colorants in Almería, al-Andalus

Article Title: Materials and technological choices in ceramic production of Medieval Almería (al-Andalus): clays, glazes and colorants

Article References: Peregrina, M. J., Molera, J., Garcia-Porras, A., Pradell, T., & Salinas, E. (2026). Materials and technological choices in ceramic production of Medieval Almería (al-Andalus): clays, glazes and colorants. Archaeological and Anthropological Sciences, 18(11), Article 219. https://doi.org/10.1007/s12520-026-02575-6

Image Credits: AI Generated

DOI: 10.1007/s12520-026-02575-6

Keywords: medieval pottery, archaeometry, Almería, al-Andalus, ceramic technology, clay fabrics, lead glazes, tin opacification, pottery workshops, X-ray diffraction, SEM–EDS, petrography