Navigation – Plan du site

AccueilNuméros135-2Le forme del vetro: tecnologie a ...The glass finds from the Vetricel...

Le forme del vetro: tecnologie a confronto. Produzioni vitree e invetriate in Sicilia, Italia peninsulare, Ifrīqiya e al-Andalus tra IX e XI secolo

The glass finds from the Vetricella site (9th-12th c.)Introduction

Archaelogy and archaeometry
Bernard Gratuze, Letizia Maria Castelli et Giovanna Bianchi
p. 349-359

Résumé

We examine here the glass finds from the Vetricella site. This site, investigated during the ERC-Avanced nEU-Med project (www.neu-med.unisi.it ) is interpreted as the centre of the royal curtis of Valli attested in documentary sources from 937. During the excavation, a large number of glass objects dated from the 4th to the 12th centuries were unearthed. The compositions of the early mediaeval Vetricella glass vessels (beakers, goblets, bottles, ampoules/vials), are studied and compared with those of similar finds from various other European sites. With the exception of one glass bottle, made with Syro-palestinian plant ash soda glass, all the other studied glasses belong to Antique or Late-Antique natron glass groups (Mn- and Sb-decolorized, Foy 2 and HIMT). All these objects illustrate the recycling practices of Antique and Late Antique glass in Europe as they are described by Theophilus for the 11th and 12th centuries.

Haut de page

Texte intégral

Introduction

  • 1 Gratuze 2020.

1Important changes in the glass manufacturing process occurred in the Mediterranean world at the end of the 8th century. This phenomenon stems directly from a progressive decline of the production of raw natron glass in the Near-East. From the 9th century onward, in Eastern Mediterranean and Mesopotamia, soda-lime glass made from halophytic plant ash becomes the prevailing type of glass before becoming by the 12th century the dominant glass type throughout the Mediterranean. However, natron glass seems to be still used until the end of the 12th century in Europe for common vessels and for more specific productions such as the cobalt blue vessels and churches window panes. In the recent years, the increasing number of analyses carried out on Western European glass dated from this transition period has highlighted both the presence of a great variety of different glass compositions and the importance of natron glass recycling. A large part of the Vetricella glass vessels (cup, beakers, goblets, bottles, ampoules/vials), unearthed on this site, belong to this transition period. The compositions of these objects, which were analysed by LA-ICP-MS at the IRAMAT, Centre Ernest-Babelon (CNRS, France) in collaboration with the nEU-Med project team, is compared with those of Antique and Late Antique glass productions as well as with those of Islamic Near Eastern glasses and of Early Middle Ages European productions. A large part of the results obtained on the cup of blue glass decorated with opaque white glass strands and pellets and red glass trails has already been published.1 Therefore, these results won’t be presented in detail here but will be mainly used as comparative data.

The Vetricella site and glass finds

2Vetricella, Scarlino (GR), is located in Tuscany, in the central portion of the coastal plain crossed by the river Pecora, at the foot of the castle of Scarlino; discovered in 2005 thanks to aerial reconnaissance: three large concentric circles that develop around a quadrangular trace.

3The site has been part part of the project: "Origins of a new economic union (7th-12th centuries): resources, landscapes and political strategies in a Mediterranean region", European project within Horizon 2020, ERC Advanced. Host institution is the University of Siena, directed by Professor Richard Hodges (P.I. of the project) and Professor Giovanna Bianchi (Unisi coordinator). It started at the end of 2015 and finished in 2021.

4The Vetricella site has become the key-study of the project due to the wealth of data collected in multidisciplinary surveys. The site is interpreted with the center of a royal Curtis, called Valli, attested for the first time in 937 AD but with a continuity of life from the 8th to the 12th century.

  • 2 Castelli 2020.

5Glass artefacts found at the site of Vetricella comprise 168 fragments.2 Identification was not possible for 50 of these. 118 of the 168 sherds have been analysed, of which 83 different forms and 4 semi-finished fragments have been identified.

6The identified glass fragments have been divided into seven different forms as well as semi-finished production (fig. 1): I. Cups/Small Cups, 2. Beakers, 3. Goblets, 4. Lamps, 5. Ampoules/vials, 6. Bottles, 7. Gaming pieces, 8. Semi-finished products (shapeless molten glass).

Figure 1. Some examples of Vetricella Glasses.

Figure 1. Some examples of Vetricella Glasses.
  • 3 Castelli 2020.

7Different types, dated from Late Antiquity to Early Middle Ages were identified for the cups, goblets, lamps and beakers.3 The large number of goblets from Vetricella (20 fragment of goblets minimus number) is evocative of similar discoveries in the Domus solarata from Forum of Nerva (Roma, later 10th or early 11th centuries).

8The study of the glass finds, the typological analysis and the discussion on the stratigraphic provenance have been extensively discussed by the writer in the article in the volume of the project. (L. Castelli, Glass artefacts from the site of Vetricella (Scarlino, Grosseto), in G. Bianchi, R. Hodges (eds.), The nEU-Med project: Vetricella, an Early Medieval royal property on Tuscany's Mediterranean, Florence, 2020, Library of Medieval Archeology 28, p. 69-75.)

The studied glass corpus

9The 43 selected samples were chosen among the different vessel types dated from Early Middles Ages levels. They consist of (fig. 2 and tab. 1):

  • 5 sherds belonging to a cup of blue glass decorated with opaque white glass strands and pellets and red glass trails and dated from 10th-12th century. Thirteen sherds from the same cup were previously analysed and published.4 The fragments of blue Glass were concentrated in the eastern area of ​​the site, near the ecclesiastical building, in contexts dating back to the mid XI century.
  • 38 greenish and colourless sherds originating from different vessel types:
    • 8 greenish sherds with some red trails originating from 4 beakers type 2 dated from Early Middles Ages,
    • 22 sherds originating from 14 goblets type 1 (mostly greenish or pale greenish, except 2 colourless samples) and 8 goblets type 2 (mostly colourless or slightly pale greenish) dated from 10th-13th century. Goblets were positioned in different areas around the tower, in stratigraphy almost all ascribable to the 10th-12th century.
    • 2 sherds belonging to two bottles dated as Early Medieval, one pale green and the other darker,
    • 4 greenish sherds from 2 ampoules and 1 vials (2 sherds) dated from 11th-12th century
    • and 2 sherds from semi-finished products (shapeless molten glass), one dark green and one greenish, possibly indicative of production activities on site.

Table 1. Studied corpus of Vetricella glasses.

US N. Schedule TYPE N. Sherds Date N. Sample Chem. group
127 19 Blue glasses Cup type 2 1 X-XII c. 1 Roman
0 7 Blue glasses Cup type 2 2 X-XII c. 2 A & B Roman
178 12 Blue glasses Cup type 2 1 X-XII c. 3 Roman
fondo trincea B 20 Blue glasses Cup type 2 1 X-XII c. 4 Roman
178 8 Beakers type 2 3 Early Middle Ages 5 A, B, C Foy 2
130 22 Beakers type 2 3 Early Middle Ages 6 A, B, C Foy 2
Q.I. 8 74 Beakers type 2 1 Early Middle Ages 7 Foy 2
3006 102 Beakers type 2 1 Early Middle Ages 8 Foy 2
0 21 Goblets type 1 1 X-XIII c. 9 Roman Mn-Sb
Q. I. 8 25 Goblets type 1 1 X-XIII c. 10 Roman Mn-Sb
ripulitura 37 Goblets type 1 1 X-XIII c. 11 Foy 2
sporadico 39 Goblets type 1 1 X-XIII c. 12 Roman Sb-Mn
385 46 Goblets type 1 1 X-XIII c. 13 HIMT/Foy 2
Q. E. 11 51 Goblets type 1 1 X-XIII c. 14 Foy 2
837 54 Goblets type 1 1 X-XIII c. 15 Foy 2
860 58 Goblets type 1 1 X-XIII c. 16 Foy 2
848 57 Goblets type 1 1 X-XIII c. 17 Foy 2
844 56 Goblets type 1 1 X-XIII c. 18 Foy 2
968 60 Goblets type 1 1 X-XIII c. 19 Foy 2
844 79 Goblets type 1 1 X-XIII c. 20 Foy 2
4001 105 Goblets type 1 1 X-XIII c. 21 Foy 2
4001 104 Goblets type 1 1 X-XIII c. 22 Foy 2
127 6 Goblets type 2 1 X-XIII c. 23 Roman Sb-Mn
ripulitura 2 Goblets type 2 1 X-XIII c. 24 Roman Sb-Mn
0 19 Goblets type 2 1 X-XIII c. 25 Roman Mn-Sb
0 28 Goblets type 2 1 X-XIII c. 26 Roman Sb-Mn
160 32 Goblets type 2 1 X-XIII c. 27 Roman Sb-Mn
0 77 Goblets type 2 1 X-XIII c. 28 Roman Sb-Mn
837 81 Goblets type 2 1 X-XIII c. 29 Roman Sb-Mn
2001 89 Goblets type 2 1 X-XIII c. 30 Roman Sb-Mn
ripulitura 36 Bottle 1 Early Middle Ages 31 Syro-palestinian soda plant ash glass
863 59 Bottle 1 Early Middle Ages 32 HIMT
127 12 Ampoules 1 XI-XII c. 33 Foy 2
ripulitura 42 Vials 2 XI-XII c. 34 A & B Foy 2
848 80 Ampoules 1 XI-XII c. 35 Foy 2
Q. I. 9 38 Semi-finished 1 36 Foy 2/HIMT
ripulitura 70 Semi-finished 1 37 Foy 2

Figure 2. LA-ICP-MS analytical cell (S155 from Resonetics) used for the analyses of the Vetricella glass corpus. Samples in the metallic cylinders are the glass standards.

Figure 2. LA-ICP-MS analytical cell (S155 from Resonetics) used for the analyses of the Vetricella glass corpus. Samples in the metallic cylinders are the glass standards.

Analytical method

  • 5 Gratuze 2016.
  • 6 Gratuze 2016.

10The analyses of the glass sherds were carried out by Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry (LA-ICP-MS). The instrumentation consisted of a Resolution M50E UV laser probe from Resonetics/ASI (Eximer ArF laser working at 193 nm equipped with the S155 cell) coupled with a Thermo Fisher Scientific ELEMENT XR mass spectrometer.5 LA-ICP-MS allows a nearly non-destructive analysis, invisible to the naked eye, of the glass objects. Analytical parameters were as follow: the excimer laser was operated at 5.5 mJ with a repetition rate of 10 Hz, ablation time was set to 50 seconds: 20 seconds pre-ablation, so that transient signal and possible surface corrosion or contamination could be removed, and 30 seconds collection time corresponding to 9 mass scans from lithium to uranium. The signal was measured in counts/second, in low resolution mode for 58 different isotopes. The fifty-eight elements include all major, minor (except sulphur) and trace elements which are usually present in glass samples.6 Blanks were run periodically between series of 20 analyses. Spot sizes were set to 100 µm (although reduced down to 70 µm when saturation occurred for element such as manganese or copper). During analysis live counts were continuously observed: when element spikes signifying the presence of inclusions were observed, results were discarded and a new site selected. From one to eight areas were analysed per sample in order to analyse the different part and different coloured glasses present on the sherds; homogeneity and agreement between runs carried out on the same type of glass were consistently good. However, some samples show variations in composition due either to the presence of red trails, with significantly different composition, within the glass or to imperfect mixing of glasses of different compositions (insufficient temperature and melting time). In order to take these variabilities into account, we have chosen to represent all the analyses performed on the graphs rather than the average compositions of the objects.

11Calibration was performed using five reference standards: NIST610, Corning B, C and D, and APL1 (an in-house reference glass used for chlorine determination), were run periodically (every 15 to 20 samples) to correct for eventual drifts. The standards are used to calculate the response coefficient (k) of each element. The measured values were normalised against 28Si, the internal standard. Concentrations are calculated assuming that the sum of the concentrations of the measured elements is equal to 100 weight percent. In total, 58 elements were recorded. For the major and minor elements accuracy and precision were within 5 % relative and within 10 % for most trace elements.

Results

Identification and provenance of glass groups

12According to their MgO (from 0.40 % to 0.96 %) and K2O (from 0.36 % to 0.93%) contents (fig. 3), with the exception of one bottle, all the analyzed glasses from Vetricella are mineral soda-lime glass, that is glass obtained by melting a siliceous sand with a mineral soda flux such as natron (fig. 3).

13One of the studied bottle (N 31) is characterized by higher MgO (2.57 %) and K2O (1.91 %) contents and was made with soda originating from halophytic plant ashes such as Salsola sp or Salicornia sp. We can notice that the second bottle (N 32) is characterized by the lowest potash (0.36 %) content and, with the exception of the white opaque glasses, the highest magnesia content (0.91 %) compared to the other natron glasses.

Figure 3. Distribution of the analysed glasses according to their magnesia and potash contents, on the left, all the glasses are represented, on the right, focus on the natron glass group. All the analytical results are plotted some glasses are represented by 2 to 5 points.

Figure 3. Distribution of the analysed glasses according to their magnesia and potash contents, on the left, all the glasses are represented, on the right, focus on the natron glass group. All the analytical results are plotted some glasses are represented by 2 to 5 points.

14A more detailed study of the potash and magnesia contents of natron glasses (fig. 3 right) shows that the different groups of objects are characterized by increasing magnesia and potash contents from the goblets type 2 to the white opaque glasses. Four main groups can be considered:

  • In the first one (I) we find all the goblets type 2 plus two goblets type 1 (N 10 and 12). These glasses have the lowest potash and magnesia contents;
  • The second group (II) contains most of the analyses carried out on the cobalt blue glass of the cup plus one goblet type 1 (N 21);
  • The third group (III) brings together all the remaining goblets type 1, most of the beakers (except two analyses carried out on the red glass of beaker 5) as well as all the analyses carried out on the red glass trails of the blue cup with some analyses carried out on the white glasses;
  • In the last group (IV), we find mainly the white opaque glass together with the analyses done on the red glass of beaker 5 and on the red glass of waste 37 as well as the two analyses carried out on the waste 36.
  • 7 Freestone 2021.

15If we refer to the classification of late Antique and early Middle Ages glasses, according to their TiO2, Al2O3 and SiO2 contents (fig. 4), the Vetricella’s glasses can be classified among the following glass groups: Foy 2, Roman Mn, Roman Sb and HIMT, that is mainly Egyptian glass groups or mixing between Egyptian and Syro-palestinian (Roman Mn) for the common vessels.7 Whit the exception of the natron bottle which belongs to the HIMT group (fig. 4), all the other natron glasses belong to Foy 2 glass group or to groups obtained by mixing Roman Mn and Sb decolourised glass, or Foy 2 and other groups (Roman Mn/Roman Sb/HIMT, fig. 5 and 6).

  • 8 Pactat – Gratuze 2022.

Figure 4. Distribution of the analysed natron glasses according to their Al2O3/SiO2 and TiO2/Al2O3 ratios. Comparison with Antique, late Antique and early Mediaeval groups of glass.8 On the right, focus on the main Vetricella’s natron glass group. All the analytical results are plotted some glasses are represented by 2 to 5 points.

Figure 4. Distribution of the analysed natron glasses according to their Al2O3/SiO2 and TiO2/Al2O3 ratios. Comparison with Antique, late Antique and early Mediaeval groups of glass.8 On the right, focus on the main Vetricella’s natron glass group. All the analytical results are plotted some glasses are represented by 2 to 5 points.

Figure 5. Distribution of the analysed glasses according to their TiO2 and ZrO2 contents. On the right, focus on the lower left part of the diagram. All the analytical results are plotted some glasses are represented by 2 to 5 points.

Figure 5. Distribution of the analysed glasses according to their TiO2 and ZrO2 contents. On the right, focus on the lower left part of the diagram. All the analytical results are plotted some glasses are represented by 2 to 5 points.

Figure 6. Distribution of the analysed glasses according to their MnO and Sb2O3 contents. The high antimony contents of all the glasses from the cobalt blue cup are probably due to the fact that these glasses originate from the recycling of ancient Roman tesserae.

Figure 6. Distribution of the analysed glasses according to their MnO and Sb2O3 contents. The high antimony contents of all the glasses from the cobalt blue cup are probably due to the fact that these glasses originate from the recycling of ancient Roman tesserae.
  • 9 Gratuze 2020.

16According to their SrO/CaO ratios, the glass of the blue cup has probably a Syro-palestinian origin (fig. 7) and, as it has already been discussed in a preceding paper9 the glasses used to make that cup originates probably from the recycling of Roman tesserae if we refer to their high contents in antimony (fig. 6). We can also notice that despite their distribution on different diagrams(fig. 4 and 5), the red glasses from the cup cannot be classified as Foy 2 glass but have probably also a Syro-palestinian origin (fig. 7).

  • 10 Phelps 2018; Schibille et al. 2019.

17On figure 7 we also observe that the plant ash glass used to make the bottle N 31 has similar SrO/CaO ratio to the natron glasses used to make the cup. It probably indicates that this glass originates from Islamic Syro-palestinian plant ash glass workshops.10

Figure 7. Distribution of the analysed glasses according to their CaO and SrO contents. With the exception of the HIMT glass, all the other glasses are distributed among two main trends. On the left, we find all the natron glass common vessels with an average SrO/CaO ratio equal to 0.0078 (probably with an Egyptian origin) and on the left the glasses of the blue cup with the bottle plant ash glass with an average SrO/CaO ratio equal to 0.0071 (probably with a Syro-palestinian origin).

Figure 7. Distribution of the analysed glasses according to their CaO and SrO contents. With the exception of the HIMT glass, all the other glasses are distributed among two main trends. On the left, we find all the natron glass common vessels with an average SrO/CaO ratio equal to 0.0078 (probably with an Egyptian origin) and on the left the glasses of the blue cup with the bottle plant ash glass with an average SrO/CaO ratio equal to 0.0071 (probably with a Syro-palestinian origin).

Red glass trails: accidental or not?

18If we consider now the red glass trails which are present in both the blue cup and the beakers, we can notice on figure 8 that the reduction of copper oxide into metallic copper originate probably from two different processes. In the cobalt blue cup, the red glass is characterized by higher iron, copper and tin contents than the blue glass. It also contains less antimony and more manganese, titanium and zirconium (fig. 5 and 6). Thus the red glasses present in the cup have completely different compositions from the blue glass and originates probably from antique red glass (may be tesserae) mixed voluntary or accidentally with the cobalt blue tesserae recycled to make that cup.

19On the contrary in the beakers, the red glass has a composition which appears fairly similar to that of the based greenish glass. It mainly contains a little bit more iron oxide, but presents similar contents of tin and copper as well as all other elements. We can therefore hypothesize that the red glass trails present in the beakers may originate from an accidental reduction, in the crucible, of the copper oxide, originally present in the glass. Its reduction into metallic copper may be due to both, the presence of an excess of iron (contaminations from tools or local heterogeneities) and favourable reductant conditions in the crucible.

Figure 8. Distribution of the analysed glasses according to their CuO and Fe2O3 contents on the left and CuO and SnO2 contents on the right. The red glass from the cup exhibits higher copper, iron and tin contents, while the red glass from the beakers exhibits mainly higher iron contents, illustrating that the reduction of copper in these red glasses was not obtained from the same colouring processes.

Figure 8. Distribution of the analysed glasses according to their CuO and Fe2O3 contents on the left and CuO and SnO2 contents on the right. The red glass from the cup exhibits higher copper, iron and tin contents, while the red glass from the beakers exhibits mainly higher iron contents, illustrating that the reduction of copper in these red glasses was not obtained from the same colouring processes.

Figure 9. Distribution of the analysed glasses according to: at the top, their CuO and SnO2 contents on the left and CuO and PbO contents on the right and at the bottom to their Cl, K2O an P2O5 contents. In colourless glasses, contents in copper, lead and tin oxides are usually used as indicator of intensity of recycling (> 0.01 %). These criteria do not apply to the cobalt blue cup glasses as these elements originate from the colouring process. Recycling tend also to increase potash and phosphorus contents (ashes contamination) and decrease contents of volatile element such as chlorine. Applying these criteria to our corpus we observe that the bottles (HIMT and soda plant ash glasses) and in a less extent goblets type 2, were made from less recycled glass than goblets type 1, beakers and ampoules/vials.

Figure 9. Distribution of the analysed glasses according to: at the top, their CuO and SnO2 contents on the left and CuO and PbO contents on the right and at the bottom to their Cl, K2O an P2O5 contents. In colourless glasses, contents in copper, lead and tin oxides are usually used as indicator of intensity of recycling (> 0.01 %). These criteria do not apply to the cobalt blue cup glasses as these elements originate from the colouring process. Recycling tend also to increase potash and phosphorus contents (ashes contamination) and decrease contents of volatile element such as chlorine. Applying these criteria to our corpus we observe that the bottles (HIMT and soda plant ash glasses) and in a less extent goblets type 2, were made from less recycled glass than goblets type 1, beakers and ampoules/vials.

Glass recycling

  • 11 Ceglia et al. 2019.
  • 12 Gratuze 2020.

20If we now try to estimate the intensity of glass recycling by using the respective contents of copper, tin and lead oxide11 in the colourless and greenish glasses (fig. 9 top), we observe that Foy 2 glasses (beakers, goblets type 1 and ampoules/vials) present the highest contents for theses oxides and were therefore probably made with the most recycled glasses. On the other hand, goblets type 2 and some goblets type 1, made with mixed Roman Mn and Sb glasses show among the lowest contents of Sn and Cu and content of lead oxides just above the accepted recycling threshold value of 100 ppm level. But the lowest contents for these three oxides were measured for the two bottle made with HIMT and soda plant ash glass. If these results are not surprising for the soda plant ash glass whose production period is probably contemporaneous of Vetricella’s occupation period, it is more puzzling for the HIMT glass whose production period dated from the 4th and 5th centuries. These results seem to show that most of the glass used by the glassmakers, who produced the tableware found at Vetricella, did not come from multiple recycling and was probably easily accessible. Part of this "fresh" glass could have come from architectural glass (window panes) recovered from ancient buildings. Of course, we cannot use the same parameters to estimate the intensity of glass recycling of the cobalt blue cup glasses as these metallic elements are part of the colouring process. For these glasses we can however study the relationship between potash, phosphorus and chlorine (fig. 9 bottom). The contents of first two elements which are present in the furnace atmosphere tend to increase accordingly to the re-melting events while the contents of Cl tend to decrease due to the volatility of this element. Similar results are obtained for all these elements for natron glasses. Goblets type 2 contain the lowest potash and phosphorus contents associated with the highest chlorine contents, while beakers which had the highest contents of metallic impurities have also the highest contents in potash and phosphorus and the lowest contents in chlorine. With these parameters, evidence of recycling is also observed for the coloured glasses of the cobalt blue cup: potash and phosphorus show a positive correlation while a negative correlation is observed between potash and chlorine. If we interpret in term of recycling events the two groups of compositions, observed among the different analysed fragments of the cup and discussed in the preceding paper,12 we may hypothesize that a sets of fresh tesserae was added into a crucible containing already melted tesserae and that the two sets were incompletely mixed during melting prior to the making of the cup.

21These criteria cannot apply to plant ash soda glasses as they contain higher and more variable amount of potash and phosphorus.

Conclusion

22With the exception of one bottle made with contemporaneous Syro-palestinian soda plant ash glass, all the other studied Vetricella’s glasses were made from recycled Antique natron glasses produced before the 7th century.

23The natron recycled glasses form two main groups:

  • On one side we have the cobalt blue molded glass cup decorated with white opaque glass strands and pellets and red glass trails. Despite a sligltly different typology, its composition is similar to the other vessels of that type found throughout Europe during Ottonian period.13 The compositions of these vessels illustrate the recycling practices of ancient coloured glass tesserae in Europe, as described by Theophilus for the 11th and 12th centuries (Chap. XII).
  • On the other side we have all the colourless or greenish vessels (goblets, beakers…). Their compositions correspond to the recycling of more common glass (windows, vessels, chunks of raw glass…) and can be classified among the more common late antique glass groups: Foy 2, Roman Mn and Sb decolorised glass and HITM. Mixing of glasses belonging to these groups was also observed. It was also noticed that Beakers and Goblets Type 1 were mainly made with Foy 2 glass while Goblets Type 2 were maily made with glass obtained by mixing Roman Mn and Sb decolorised glass.
  • 14 Pactat – Gratuze 2022; Gratuze 2020.

24From the different collected information obtained on these vessels (typologies, chemical compositions, discovery context), we can hypothesize that goblets probably originate from local Italian production while the blue glass cup originate more probably from Northern European production centers.14

25As it was shown by other types of finds found in the site, Vetricella was the center of a royal court inserted in a large-scale exchange system. The importation of the blue glass cup from northern European production center fits well in this context. Within this frame, we may also reconsider the presence of the Syro-palestinian soda plant ash glass bottle as resulting from the import of Near Eastern goods of which the bottle would have been the container.

Haut de page

Bibliographie

Castelli 2020 = L. Castelli, Glass artefacts from the site of Vetricella (Scarlino, Grosseto), in G. Bianchi, R. Hodges (eds), The nEU-Med project: Vetricella, an Early Medieval royal property on Tuscany's Mediterranean, Florence, 2020 (Biblioteca di Archeologia medievale, 28), p. 69-75.

Ceglia et al. 2019 = A. Ceglia, P. Cosyns, N. Schibille, W. Meulebroeck, Unravelling provenance and recycling of Late Antique glass with trace elements, in Archaeological and Anthropological Sciences, 11, 2019, p. 279-291.

Freestone 2021 = I.C. Freestone, Glass production in the first millennium CE: a compositional perspective, in F. Klimscha, H.J. Karlsen, S. Hansen, J. Renn (ed.) Künstlichen Stein zum durchsichtigen Massenprodukt / From artificial stone to translucent mass-product. Innovationen in der Glastechnik und ihre sozialen Folgen zwischen Bronzezeit und Antike / Innovations in the Technologies of Glass and Their Social Consequences between Bronze Age and Antiquity, Berlin, 2021 (Berlin Studies of the Ancient World, 67), p. 245-263.

Gratuze 2016 = B. Gratuze, Glass Characterization Using Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry Methods, in L. Dussubieux, M. Golitko B. Gratuze (eds), Recent Advances in Laser Ablation ICP-MS for Archaeology, Berlin-Heidelberg, 2016, p. 179-196.

Gratuze 2020 = B. Gratuze, The blue and bluish green glass sherds, decorated with opaque white glass strands, discovered at Vetricella (Scarlino, Grosseto): Analytical Study, in G. Bianchi, R. Hodges (eds), The nEU-Med project: Vetricella, an Early Medieval royal property on Tuscany's Mediterranean, Florence, 2020 (Biblioteca di Archeologia medievale, 28), p. 77-88.

Pactat – Gratuze 2022 = I. Pactat, B. Gratuze, The bowl glass sherds of Saint-Savin type discovered at Haithabu: analytical study, in V. Hilberg (ed.), Haithabu 983-1066. Der Untergang eines dänischen Handelszentrums in der späten Wikingerzeit, München, 2022 (Die Ausgrabungen in Haithabu, 19, 2), Band 1, p. 457-468.

Phelps 2018 = M. Phelps, Glass supply and trade in early Islamic Ramla: an investigation of the plant ash glass, in D. Rosenow, M. Phelps, A. Meek, I.C. Freestone (eds), Things that Travelled: Mediterranean Glass in the First Millennium CE, London, 2018, p. 236-282.

Schibille et al. 2018 = N. Schibille, B. Gratuze, E. Ollivier, E. Blondeau, Chronology of early Islamic glass compositions from Egypt, in Journal of Archaeological Science, 104, 2019, p. 10-18.

Haut de page

Notes

1 Gratuze 2020.

2 Castelli 2020.

3 Castelli 2020.

4 Gratuze 2020.

5 Gratuze 2016.

6 Gratuze 2016.

7 Freestone 2021.

8 Pactat – Gratuze 2022.

9 Gratuze 2020.

10 Phelps 2018; Schibille et al. 2019.

11 Ceglia et al. 2019.

12 Gratuze 2020.

13 Pactat – Gratuze 2022; Gratuze 2020.

14 Pactat – Gratuze 2022; Gratuze 2020.

Haut de page

Table des illustrations

Titre Figure 1. Some examples of Vetricella Glasses.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-1.jpg
Fichier image/jpeg, 63k
Titre Figure 2. LA-ICP-MS analytical cell (S155 from Resonetics) used for the analyses of the Vetricella glass corpus. Samples in the metallic cylinders are the glass standards.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-2.jpg
Fichier image/jpeg, 154k
Titre Figure 3. Distribution of the analysed glasses according to their magnesia and potash contents, on the left, all the glasses are represented, on the right, focus on the natron glass group. All the analytical results are plotted some glasses are represented by 2 to 5 points.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-3.jpg
Fichier image/jpeg, 80k
Titre Figure 4. Distribution of the analysed natron glasses according to their Al2O3/SiO2 and TiO2/Al2O3 ratios. Comparison with Antique, late Antique and early Mediaeval groups of glass.8 On the right, focus on the main Vetricella’s natron glass group. All the analytical results are plotted some glasses are represented by 2 to 5 points.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-4.jpg
Fichier image/jpeg, 125k
Titre Figure 5. Distribution of the analysed glasses according to their TiO2 and ZrO2 contents. On the right, focus on the lower left part of the diagram. All the analytical results are plotted some glasses are represented by 2 to 5 points.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-5.jpg
Fichier image/jpeg, 76k
Titre Figure 6. Distribution of the analysed glasses according to their MnO and Sb2O3 contents. The high antimony contents of all the glasses from the cobalt blue cup are probably due to the fact that these glasses originate from the recycling of ancient Roman tesserae.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-6.jpg
Fichier image/jpeg, 104k
Titre Figure 7. Distribution of the analysed glasses according to their CaO and SrO contents. With the exception of the HIMT glass, all the other glasses are distributed among two main trends. On the left, we find all the natron glass common vessels with an average SrO/CaO ratio equal to 0.0078 (probably with an Egyptian origin) and on the left the glasses of the blue cup with the bottle plant ash glass with an average SrO/CaO ratio equal to 0.0071 (probably with a Syro-palestinian origin).
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-7.jpg
Fichier image/jpeg, 118k
Titre Figure 8. Distribution of the analysed glasses according to their CuO and Fe2O3 contents on the left and CuO and SnO2 contents on the right. The red glass from the cup exhibits higher copper, iron and tin contents, while the red glass from the beakers exhibits mainly higher iron contents, illustrating that the reduction of copper in these red glasses was not obtained from the same colouring processes.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-8.jpg
Fichier image/jpeg, 73k
Titre Figure 9. Distribution of the analysed glasses according to: at the top, their CuO and SnO2 contents on the left and CuO and PbO contents on the right and at the bottom to their Cl, K2O an P2O5 contents. In colourless glasses, contents in copper, lead and tin oxides are usually used as indicator of intensity of recycling (> 0.01 %). These criteria do not apply to the cobalt blue cup glasses as these elements originate from the colouring process. Recycling tend also to increase potash and phosphorus contents (ashes contamination) and decrease contents of volatile element such as chlorine. Applying these criteria to our corpus we observe that the bottles (HIMT and soda plant ash glasses) and in a less extent goblets type 2, were made from less recycled glass than goblets type 1, beakers and ampoules/vials.
URL http://journals.openedition.org/mefrm/docannexe/image/12631/img-9.jpg
Fichier image/jpeg, 114k
Haut de page

Pour citer cet article

Référence papier

Bernard Gratuze, Letizia Maria Castelli et Giovanna Bianchi, « The glass finds from the Vetricella site (9th-12th c.)Introduction »Mélanges de l’École française de Rome - Moyen Âge, 135-2 | 2023, 349-359.

Référence électronique

Bernard Gratuze, Letizia Maria Castelli et Giovanna Bianchi, « The glass finds from the Vetricella site (9th-12th c.)Introduction »Mélanges de l’École française de Rome - Moyen Âge [En ligne], 135-2 | 2023, mis en ligne le 01 mars 2024, consulté le 18 mars 2025. URL : http://journals.openedition.org/mefrm/12631 ; DOI : https://doi.org/10.4000/mefrm.12631

Haut de page

Auteurs

Bernard Gratuze

Institut de Recherche sur les Archéomatériaux/Centre Ernest-Babelon, UMR7065, CNRS/Université d’Orléans, bernard.gratuze@cnrs-orleans.fr

Letizia Maria Castelli

Università degli Studi di Siena, letiziamaria.castelli@gmail.com

Giovanna Bianchi

Università degli Studi di Siena, giovanna.bianchi@unisi.it

Haut de page

Droits d’auteur

CC-BY-NC-ND-4.0

Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search