Fig 1.
Archaeological sites in West Africa.
Location of the archaeological sites where the analysed glass beads were found, as well as of other West African archaeological sites mentioned in this paper (map: courtesy of CIA’s The World Factbook 2020, modified).
Fig 2.
Glass beads found in Dourou-Boro, Mali.
Table 1.
Morphological and visual characteristics of the 21 glass beads found in Dourou-Boro and Sadia (Mali), and in Djoutoubaya (Senegal).
Fig 3.
Glass beads found in Sadia, Mali.
The two deteriorated beads not subjected to chemical analysis are indicated with “*”.
Fig 4.
Glass beads found in Djoutoubaya, Senegal.
The three deteriorated beads not subjected to chemical analysis are indicated with “*”.
Table 2.
Reduced composition (*) of the analysed glass beads by site.
Fig 5.
Sample grouping based on trace element content.
Trace elements trends from the analysed samples revealing various subgroups within the two main clusters.
Table 3.
Chemical composition of the glass beads from Dourou-Boro, Djoutoubaya, and Sadia (cf. Table 1 for abbreviations).
Fig 6.
Subdivision of the low lime and magnesia samples based on trace element content.
Fig 7.
Concentrations of the elements linked to the flux in the analysed glass beads.
Boron versus lithium (A) and phosphorus pentoxide versus magnesia (B) suggest different origins for the plant ash used in production.
Fig 8.
Subdivision of the high magnesia samples based on trace element content.
Fig 9.
Different origins of the turquoise beads copper colorants.
Zinc to copper (A) and nickel to copper (B) show that beads DB-9t, DB-11t, and DJ16-5t have been coloured with the same type of colouring agent, which is different from those used for the other turquoise beads.
Fig 10.
Location of the archaeological sites where the plant-ash glasses used as reference were found (map: Courtesy of CIA’s The World Factbook 2020, modified).
Fig 11.
Dourou-Boro, Djoutoubaya, and Sadia glasses compared to contemporary glass assemblages.
Alumina versus magnesia to lime ratios of published data of glasses from Levantine and Mesopotamian contemporary sites, grouped after Phelps [49, 50].
Fig 12.
Trace elements composition of Dourou-Boro and Djoutoubaya glasses compared to contemporary glasses.
Zirconium to titanium versus chromium to lanthanum ratios of glasses from Dourou-Boro and Djoutoubaya compared to Levantine and Mesopotamian glasses showing the different sources of raw materials.
Fig 13.
Dourou-Boro and Djoutoubaya glasses compared to Egyptian, Levantine, and Mesopotamian glasses.
Titanium oxide to alumina ratio versus magnesia (A) and zirconium versus titanium (B) confirm the probable different origin of the glasses from Dourou-Boro and Djoutoubaya.
Fig 14.
The Sadia glass bead compared to contemporary high alumina glasses.
Lime versus the sum of soda, magnesia and potash (A, C) and titanium oxide versus alumina (B, D) of high alumina glass from Sadia compared to contemporary glasses from the Iberian Peninsula (A, B) and from eastern and southern Africa (C, D).
Fig 15.
Trace element composition of the Sadia bead compared to contemporary high alumina glasses.
Cerium versus zirconium (A) and uranium versus barium (B) of the Sadia bead compared to Iberian, southern, and eastern African glasses show compatibility with group 5 glass from Ciudad de Vascos and with South African Mapungubwe Oblate.
Table 4.
West African glass bead assemblages considered for comparison.
Fig 16.
Comparison between contemporary glass bead assemblages from West Africa.
Alumina versus magnesia to lime ratios of published compositions of glass beads from West Africa, grouped after Phelps [49, 50].